Daily Papers

The inference overhead induced by redundant reasoning undermines the interactive experience and severely bottlenecks the deployment of Large Reasoning Models. Existing reinforcement learning (RL)-based solutions tackle this problem by coupling a length penalty with outcome-based rewards. This simplistic reward weighting struggles to reconcile brevity with accuracy, as enforcing brevity may compromise critical reasoning logic. In this work, we address this limitation by proposing a multi-agent RL framework that selectively penalizes redundant chunks, while preserving essential reasoning logic. Our framework, Self-Compression via MARL (SCMA), instantiates redundancy detection and evaluation through two specialized agents: a Segmentation Agent for decomposing the reasoning process into logical chunks, and a Scoring Agent for quantifying the significance of each chunk. The Segmentation and Scoring agents collaboratively define an importance-weighted length penalty during training, incentivizing a Reasoning Agent to prioritize essential logic without introducing inference overhead during deployment. Empirical evaluations across model scales demonstrate that SCMA reduces response length by 11.1\% to 39.0\% while boosting accuracy by 4.33\% to 10.02\%. Furthermore, ablation studies and qualitative analysis validate that the synergistic optimization within the MARL framework fosters emergent behaviors, yielding more powerful LRMs compared to vanilla RL paradigms.

Automatic medical report generation supports clinical diagnosis, reduces the workload of radiologists, and holds the promise of improving diagnosis consistency. However, existing evaluation metrics primarily assess the accuracy of key medical information coverage in generated reports compared to human-written reports, while overlooking crucial details such as the location and certainty of reported abnormalities. These limitations hinder the comprehensive assessment of the reliability of generated reports and pose risks in their selection for clinical use. Therefore, we propose a Granular Explainable Multi-Agent Score (GEMA-Score) in this paper, which conducts both objective quantification and subjective evaluation through a large language model-based multi-agent workflow. Our GEMA-Score parses structured reports and employs NER-F1 calculations through interactive exchanges of information among agents to assess disease diagnosis, location, severity, and uncertainty. Additionally, an LLM-based scoring agent evaluates completeness, readability, and clinical terminology while providing explanatory feedback. Extensive experiments validate that GEMA-Score achieves the highest correlation with human expert evaluations on a public dataset, demonstrating its effectiveness in clinical scoring (Kendall coefficient = 0.70 for Rexval dataset and Kendall coefficient = 0.54 for RadEvalX dataset). The anonymous project demo is available at: https://github.com/Zhenxuan-Zhang/GEMA_score.

In this paper, we present AgentDisCo, a novel Disentangled and Collaborative agentic architecture that formulates deep research as an adversarial optimization problem between information exploration and exploitation. Unlike existing approaches that conflate these two processes into a single module, AgentDisCo employs a critic agent to evaluate generated outlines and refine search queries, and a generator agent to retrieve updated results and revise outlines accordingly. The iteratively refined outline is then passed to a downstream report writer that synthesizes a comprehensive research report. The overall workflow supports both handcrafted and automatically discovered design strategies via a meta-optimization harness, in which the generator agent is repurposed as a scoring agent to evaluate critic outputs and generate quality signals. Powerful code-generation agents (e.g., Claude-Code, Codex) systematically explore agent configurations and construct a policy bank, a structured repository of reusable design strategies, enabling the framework to self-refine without extensive human intervention. We evaluate AgentDisCo on three established deep research benchmarks (DeepResearchBench, DeepConsult, DeepResearchGym) using Gemini-2.5-Pro, achieving performance comparable to or surpassing leading closed-source systems. Observing that existing benchmarks inadequately reflect real-world user needs, we introduce GALA (General AI Life Assistants), a benchmark that mines latent research interests from users' historical browsing behavior. We further develop a rendering agent that converts research reports into visually rich poster presentations, and demonstrate an end-to-end product, AutoResearch Your Interest, which delivers personalized deep research recommendations derived from individual browsing histories.

Generative AI is being leveraged to solve a variety of computer-use tasks involving desktop applications. State-of-the-art systems have focused solely on improving accuracy on leading benchmarks. However, these systems are practically unusable due to extremely high end-to-end latency (e.g., tens of minutes) for tasks that typically take humans just a few minutes to complete. To understand the cause behind this and to guide future developments of computer agents, we conduct the first study on the temporal performance of computer-use agents on OSWorld, the flagship benchmark in computer-use AI. We find that large model calls for planning and reflection account for the majority of the overall latency, and as an agent uses more steps to complete a task, each successive step can take 3x longer than steps at the beginning of a task. We then construct OSWorld-Human, a manually annotated version of the original OSWorld dataset that contains a human-determined trajectory for each task. We evaluate 16 agents on their efficiency using OSWorld-Human and found that even the highest-scoring agents on OSWorld take 1.4-2.7x more steps than necessary.

Stronger code agents are commonly assumed to be superior teachers for post-training, yet this assumption remains poorly disentangled from task difficulty, harness design, and student capacity. We investigate this pedagogical link using Terminal-Lego, a scalable pipeline that transforms multi-domain real-world issues into environment-verified agentic tasks. Surprisingly, standalone performance does not dictate teaching efficacy: while Claude Opus 4.6 achieves higher scores on Terminal-Bench 2.0, students fine-tuned on trajectories from DeepSeek-V3.2, a lower-scoring agent, exhibit significantly stronger generalization. We attribute this "pedagogical paradox" to Environment-Grounded Supervision (EGS): trajectories that explicitly expose inspect-act-verify behaviors through harness-visible interactions allow students to internalize robust problem-solving routines rather than fragile action sequences. Scaling analysis reveals exceptional data efficiency: with only 15.3k Terminal-Lego trajectories, for example, Qwen3-32B achieves a 24.3% score on Terminal-Bench 2.0, rivaling previous SOTA performance established with over 30x the data volume. Our results suggest that the frontier of agent post-training lies beyond mere outcome-matching, shifting the focus toward "Harness Engineering", where the systematic design of environment-grounded interaction structures serves as the primary catalyst for reproducible and generalizable agentic intelligence.

Objective. Clinical AI documentation systems require evaluation methodologies that are clinically valid, economically viable, and sensitive to iterative changes. Methods requiring expert review per scoring instance are too slow and expensive for safe, iterative deployment. We present a case-specific, clinician-authored rubric methodology for clinical AI evaluation and examine whether LLM-generated rubrics can approximate clinician agreement. Materials and Methods. Twenty clinicians authored 1,646 rubrics for 823 clinical cases (736 real-world, 87 synthetic) across primary care, psychiatry, oncology, and behavioral health. Each rubric was validated by confirming that an LLM-based scoring agent consistently scored clinician-preferred outputs higher than rejected ones. Seven versions of an EHR-embedded AI agent for clinicians were evaluated across all cases. Results. Clinician-authored rubrics discriminated effectively between high- and low-quality outputs (median score gap: 82.9%) with high scoring stability (median range: 0.00%). Median scores improved from 84% to 95%. In later experiments, clinician-LLM ranking agreement (tau: 0.42-0.46) matched or exceeded clinician-clinician agreement (tau: 0.38-0.43), attributable to both ceiling compression and LLM rubric improvement. Discussion. This convergence supports incorporating LLM rubrics alongside clinician-authored ones. At roughly 1,000 times lower cost, LLM rubrics enable substantially greater evaluation coverage, while continued clinical authorship grounds evaluation in expert judgment. Ceiling compression poses a methodological challenge for future inter-rater agreement studies. Conclusion. Case-specific rubrics offer a path for clinical AI evaluation that preserves expert judgment while enabling automation at three orders lower cost. Clinician-authored rubrics establish the baseline against which LLM rubrics are validated.

Medical report interpretation plays a crucial role in healthcare, enabling both patient-facing explanations and effective information flow across clinical systems. While recent vision-language models (VLMs) and large language models (LLMs) have demonstrated general document understanding capabilities, there remains a lack of standardized benchmarks to assess structured interpretation quality in medical reports. We introduce MedRepBench, a comprehensive benchmark built from 1,900 de-identified real-world Chinese medical reports spanning diverse departments, patient demographics, and acquisition formats. The benchmark is designed primarily to evaluate end-to-end VLMs for structured medical report understanding. To enable controlled comparisons, we also include a text-only evaluation setting using high-quality OCR outputs combined with LLMs, allowing us to estimate the upper-bound performance when character recognition errors are minimized. Our evaluation framework supports two complementary protocols: (1) an objective evaluation measuring field-level recall of structured clinical items, and (2) an automated subjective evaluation using a powerful LLM as a scoring agent to assess factuality, interpretability, and reasoning quality. Based on the objective metric, we further design a reward function and apply Group Relative Policy Optimization (GRPO) to improve a mid-scale VLM, achieving up to 6% recall gain. We also observe that the OCR+LLM pipeline, despite strong performance, suffers from layout-blindness and latency issues, motivating further progress toward robust, fully vision-based report understanding.

Rubric-based scoring has become a widely used paradigm in model evaluation, typically with LLM-as-a-Judge (LaaJ) for rubric scoring. However, the reliability of LaaJ for rubric scoring remains underexplored. This concern is especially pronounced in agentic scenarios, where long, complex outputs further challenge reliable scoring. To address this, we conduct a systematic meta-evaluation of LaaJ reliability for rubric verification. We introduce RuVerBench, the first benchmark for assessing LaaJ reliability in rubric verification for agentic scenarios. RuVerBench covers two prevalent agentic domains, deep research and agentic coding, with 2,458 instances, each containing a model-generated output, a rubric, and a human-annotated label indicating whether the output satisfies the rubric. Using RuVerBench, we evaluate numerous frontier LLMs and find that even the most advanced models achieve strong performance but still exhibit substantial noise. We further analyze the impact of key LaaJ strategies, including prompt design, batching, and majority voting, on rubric verification. We find that weaker models are more sensitive to prompt variations, batched verification presents a trade-off between accuracy and efficiency, and majority voting yields effective but diminishing returns. We have released our dataset and code to facilitate future research: https://github.com/THU-KEG/RuVerBench.

Large language models (LLMs) now mediate many web-based mental-health, crisis, and other emotionally sensitive services, yet their psychosocial safety in these settings remains poorly understood and weakly evaluated. We present DialogGuard, a multi-agent framework for assessing psychosocial risks in LLM-generated responses along five high-severity dimensions: privacy violations, discriminatory behaviour, mental manipulation, psychological harm, and insulting behaviour. DialogGuard can be applied to diverse generative models through four LLM-as-a-judge pipelines, including single-agent scoring, dual-agent correction, multi-agent debate, and stochastic majority voting, grounded in a shared three-level rubric usable by both human annotators and LLM judges. Using PKU-SafeRLHF with human safety annotations, we show that multi-agent mechanisms detect psychosocial risks more accurately than non-LLM baselines and single-agent judging; dual-agent correction and majority voting provide the best trade-off between accuracy, alignment with human ratings, and robustness, while debate attains higher recall but over-flags borderline cases. We release Dialog-Guard as open-source software with a web interface that provides per-dimension risk scores and explainable natural-language rationales. A formative study with 12 practitioners illustrates how it supports prompt design, auditing, and supervision of web-facing applications for vulnerable users.

As natural language generation (NLG) models have become prevalent, systematically assessing the quality of machine-generated texts has become increasingly important. Recent studies introduce LLM-based evaluators that operate as reference-free metrics, demonstrating their capability to adeptly handle novel tasks. However, these models generally rely on a single-agent approach, which, we argue, introduces an inherent limit to their performance. This is because there exist biases in LLM agent's responses, including preferences for certain text structure or content. In this work, we propose DEBATE, an NLG evaluation framework based on multi-agent scoring system augmented with a concept of Devil's Advocate. Within the framework, one agent is instructed to criticize other agents' arguments, potentially resolving the bias in LLM agent's answers. DEBATE substantially outperforms the previous state-of-the-art methods in two meta-evaluation benchmarks in NLG evaluation, SummEval and TopicalChat. We also show that the extensiveness of debates among agents and the persona of an agent can influence the performance of evaluators.

Trustworthy deployment of LLM-based agents in software systems requires evaluating how they perform on application-specific workflows, with enough granularity to localize where they succeed and fail. Yet existing agent evaluation mechanisms are limited: benchmarks have low construct validity for application-specific workflows and environments, and replica evaluation environments are expensive and prone to drift. We propose Copy-on-Write (CoW) Scoring, a framework that evaluates agent operations directly within application environments using a PostgreSQL-level Copy-on-Write mechanism to isolate agent writes. CoW Scoring produces session- and operation-level scores that highlight where agents' database write operations succeed and fail in a given application environment, enabling inexpensive evaluation and iteration on agent harnesses and tool surfaces. We demonstrate the framework on Plane, an open-source project-management platform, where analysis surfaced specific issues in the tool surface, and corresponding fixes produced measurable improvements on affected models. Python library: https://github.com/trail-ml/agent-cow-python

The robustness of LLMs to jailbreak attacks, where users design prompts to circumvent safety measures and misuse model capabilities, has been studied primarily for LLMs acting as simple chatbots. Meanwhile, LLM agents -- which use external tools and can execute multi-stage tasks -- may pose a greater risk if misused, but their robustness remains underexplored. To facilitate research on LLM agent misuse, we propose a new benchmark called AgentHarm. The benchmark includes a diverse set of 110 explicitly malicious agent tasks (440 with augmentations), covering 11 harm categories including fraud, cybercrime, and harassment. In addition to measuring whether models refuse harmful agentic requests, scoring well on AgentHarm requires jailbroken agents to maintain their capabilities following an attack to complete a multi-step task. We evaluate a range of leading LLMs, and find (1) leading LLMs are surprisingly compliant with malicious agent requests without jailbreaking, (2) simple universal jailbreak templates can be adapted to effectively jailbreak agents, and (3) these jailbreaks enable coherent and malicious multi-step agent behavior and retain model capabilities. We publicly release AgentHarm to enable simple and reliable evaluation of attacks and defenses for LLM-based agents. We publicly release the benchmark at https://huggingface.co/ai-safety-institute/AgentHarm.

As large language models from diverse providers converge toward comparable benchmark performance, the traditional paradigm of selecting a single best model per task yields diminishing returns. We argue that orchestration topology -- the structural composition of how multiple agents are coordinated, parallelized, and synthesized -- now dominates system-level performance over individual model capability. We present AdaptOrch, a formal framework for task-adaptive multi-agent orchestration that dynamically selects among four canonical topologies (parallel, sequential, hierarchical, and hybrid) based on task dependency graphs and empirically derived domain characteristics. Our framework introduces three key contributions: (1) a Performance Convergence Scaling Law, formalizing conditions under which orchestration selection outweighs model selection; (2) a Topology Routing Algorithm that maps task decomposition DAGs to optimal orchestration patterns in O(|V| + |E|) time; and (3) an Adaptive Synthesis Protocol with provable termination guarantees and heuristic consistency scoring for parallel agent outputs. We validate AdaptOrch across coding (SWE-bench), reasoning (GPQA), and retrieval-augmented generation tasks, demonstrating that topology-aware orchestration achieves 12-23% improvement over static single-topology baselines, even when using identical underlying models. Our results establish orchestration design as a first-class optimization target independent of model scaling.

While AI agents show potential in scientific ideation, most existing frameworks rely on single-agent refinement, limiting creativity due to bounded knowledge and perspective. Inspired by real-world research dynamics, this paper investigates whether structured multi-agent discussions can surpass solitary ideation. We propose a cooperative multi-agent framework for generating research proposals and systematically compare configurations including group size, leaderled versus leaderless structures, and team compositions varying in interdisciplinarity and seniority. To assess idea quality, we employ a comprehensive protocol with agent-based scoring and human review across dimensions such as novelty, strategic vision, and integration depth. Our results show that multi-agent discussions substantially outperform solitary baselines. A designated leader acts as a catalyst, transforming discussion into more integrated and visionary proposals. Notably, we find that cognitive diversity is a primary driver of quality, yet expertise is a non-negotiable prerequisite, as teams lacking a foundation of senior knowledge fail to surpass even a single competent agent. These findings offer actionable insights for designing collaborative AI ideation systems and shed light on how team structure influences creative outcomes.

Modern retrieval pipelines increasingly serve downstream consumers like retrieval-augmented generation (RAG) and autonomous agents that need more than a scalar relevance score. A reranker that only tells the caller "how relevant" forces the agent to dump entire documents into the language-model context, wasting tokens on tangential passages and boilerplate. We introduce Prism-Reranker, a family of reranker models built on Qwen3.5 at four sizes (0.8B, 2B, 4B, 9B) that goes beyond scalar scoring. In addition to the standard yes/no relevance judgement, whenever the verdict is yes the model emits (i) a contribution statement summarizing how the document helps the query, and (ii) an evidence passage: a self-contained rewrite that preserves every query-relevant signal while discarding noise. Prism-Reranker is trained with a hybrid objective combining point-wise distillation from a strong commercial reranker API with supervised fine-tuning on contribution and evidence targets. We curate training data from KaLM-Embedding's open-source aggregation, augmented with real web documents retrieved via commercial search APIs for open-domain queries and LLM-synthesized variants, and rewrite a portion of queries into keyword-style reformulations to adapt the model to agent-issued traffic. To reconcile inconsistent labels across open corpora and obtain crisp binary supervision, we relabel data with an LLM-as-Judge ensemble aggregating votes from five frontier LLMs. On a QA subset of BEIR and on an LLM-judged evaluation of contribution and evidence quality, Prism-Reranker attains solid results across all four sizes. We further show that the same recipe extends existing LLM-based rerankers, augmenting Qwen3-Reranker-4B with contribution and evidence capabilities while improving its average BEIR-QA NDCG@10 by +1.54 over the base model. Model weights, training recipe, and evaluation suite are released.

We study the incentivized information acquisition problem, where a principal hires an agent to gather information on her behalf. Such a problem is modeled as a Stackelberg game between the principal and the agent, where the principal announces a scoring rule that specifies the payment, and then the agent then chooses an effort level that maximizes her own profit and reports the information. We study the online setting of such a problem from the principal's perspective, i.e., designing the optimal scoring rule by repeatedly interacting with the strategic agent. We design a provably sample efficient algorithm that tailors the UCB algorithm (Auer et al., 2002) to our model, which achieves a sublinear T^{2/3}-regret after T iterations. Our algorithm features a delicate estimation procedure for the optimal profit of the principal, and a conservative correction scheme that ensures the desired agent's actions are incentivized. Furthermore, a key feature of our regret bound is that it is independent of the number of states of the environment.

Tool-using LLM agents are mostly evaluated assuming all tools work. When a tool times out, returns a week-stale value, or has its description poisoned in deployment, the developer needs a controlled way to reproduce the failure, test a fix, and confirm the fix worked before deployment. We present AgentCheck, an open-source web workbench that turns an MCP server into an intervention surface. AgentCheck runs an agent against its real tools and records every tool response, then re-runs the agent with the response perturbed by a fault (12 types) injector. Matching tool calls are replayed from cache, and later tool calls go live after the agent diverges. This yields a reproduce-intervene-confirm loop: the developer toggles a mitigation, re-runs against the identical fault, and sees if the failure goes away. Scoring has two parts: deterministic pass/fail rules, plus an LLM judge for interpretive labels, validated against human annotations. Across five agents, the best passes 105/120 scenarios and the weakest only 77. The failures are usually silent, confident use of incorrect tool outputs rather than crashes. On the weakest agent, a retry mitigation raises success on timeout error faults from as few as 30% of cases to 100%, whereas stale-data faults remain near 3-4 of 10 regardless of the mitigation. AgentCheck makes these failure modes reproducible, comparable, and verifiable before deployment.

Long-horizon LLM agents can fail quietly: they settle on one reading of the evidence early, then spend the rest of the run defending it. We call this premature commitment. Final-answer scoring misses the failure mode because it sees only the answer, not whether the process has already collapsed to a stable path. We define representational commitment as cross-run hidden-state convergence at a fixed reasoning step, and use it as an early diagnostic of trajectory consistency. On Llama-3.1-70B running ReAct on HotpotQA, step-4 hidden-state similarity predicts downstream behavioral consistency (r = -0.35, partial r = -0.45), with a localized temporal and layer-wise signature. The signal replicates across Qwen-2.5-72B and Phi-3-14B, and on StrategyQA (r = -0.83). It does not track correctness: committed-wrong and committed-correct questions are not separable in activation similarity. That boundary is central to the claim. Commitment tells us whether an agent has settled, not whether it is right. A runtime monitor detects inconsistent trajectories from hidden states at AUROC up to 0.97 (0.85--0.88 under a stricter split), and a prompting intervention cuts behavioral variance by 28% against a token-matched control while leaving accuracy statistically unchanged. We also test whether the signal can route self-consistency compute; on a harder benchmark it helps only modestly and is matched by a simpler output-based baseline. The result is a diagnostic for a hidden process failure, with clear limits rather than a general accuracy lever.

While agents are increasingly spending more resources, today agent cost is mostly measured only after execution. A Budget-Aware Agent (BAGEN) should treat budget as an active control signal, rather than a passive cost metric. We first systematically define budget estimation as internal budgets (from agent computation) and external budgets (from agent actions). We then formalize budget-awareness as progressive interval estimation: at each step of a plan, an agent should predict an upper and lower bound on remaining budget, and alert when completion is unlikely. Scoring with a rollout-replay protocol, we find consistent failure patterns on four environments and five frontier agents: (1) strong agents do not necessarily have strong budget-awareness, with correlation r=0.35. (2) frontier models are consistently over-optimistic, continue spending on tasks that are unlikely to succeed, instead of alerting the user early. (3) budget-aware signal is actionable and trainable. Early stop saves 28-64% tokens on failed trajectories, and SFT+RL strengthens early stop and alert behavior. (4) precise interval calibration remains challenging, with interval coverage capping at 47% after SFT+RL. Project page: https://ragen-ai.github.io/bagen/

We investigate the mechanism design problem faced by a principal who hires multiple agents to gather and report costly information. Then, the principal exploits the information to make an informed decision. We model this problem as a game, where the principal announces a mechanism consisting in action recommendations and a payment function, a.k.a. scoring rule. Then, each agent chooses an effort level and receives partial information about an underlying state of nature based on the effort. Finally, the agents report the information (possibly non-truthfully), the principal takes a decision based on this information, and the agents are paid according to the scoring rule. While previous work focuses on single-agent problems, we consider multi-agents settings. This poses the challenge of coordinating the agents' efforts and aggregating correlated information. Indeed, we show that optimal mechanisms must correlate agents' efforts, which introduces externalities among the agents, and hence complex incentive compatibility constraints and equilibrium selection problems. First, we design a polynomial-time algorithm to find an optimal incentive compatible mechanism. Then, we study an online problem, where the principal repeatedly interacts with a group of unknown agents. We design a no-regret algorithm that provides mathcal{O}(T^{2/3}) regret with respect to an optimal mechanism, matching the state-of-the-art bound for single-agent settings.

In financial trading, large language model (LLM)-based agents demonstrate significant potential. However, the high sensitivity to market noise undermines the performance of LLM-based trading systems. To address this limitation, we propose a novel multi-agent system featuring an internal competitive mechanism inspired by modern corporate management structures. The system consists of two specialized teams: (1) Data Team - responsible for processing and condensing massive market data into diversified text factors, ensuring they fit the model's constrained context. (2) Research Team - tasked with making parallelized multipath trading decisions based on deep research methods. The core innovation lies in implementing a real-time evaluation and ranking mechanism within each team, driven by authentic market feedback. Each agent's performance undergoes continuous scoring and ranking, with only outputs from top-performing agents being adopted. The design enables the system to adaptively adjust to dynamic environment, enhances robustness against market noise and ultimately delivers superior trading performance. Experimental results demonstrate that our proposed system significantly outperforms prevailing multi-agent systems and traditional quantitative investment methods across diverse evaluation metrics. ContestTrade is open-sourced on GitHub at https://github.com/FinStep-AI/ContestTrade.

In agentic scientific machine learning (SciML), large language model (LLM) agents can discover surrogate models and select one by an automated score, typically an error metric. A low error, however, does not establish that the predicted fields satisfy the physics that matter for mechanics, such as boundary conditions, superposition, stiffness scaling, or causality. We introduce Physics-Audited Agentic SciML (PA-SciML), a verification-first workflow for agentic SciML discovery. The workflow fixes a scoring evaluator before search, derives reviewable machine-checkable physics requirements, checks each trained candidate on its outputs, and separately searches prescribed input ranges or measured load-history spans for high-violation cases without reference solution fields. A surrogate is reported as verified only under the stated checks. When enabled, the workflow also adds advisory numerical probes before training and tests one modeling change at a time to record which isolated edits are associated with score gains before reuse. In the reported computational-solid-mechanics numerical examples, the static elasticity run selects a surrogate with lower validation error than the error-only baseline while both selected models pass the common linear-elastic checks. In the transient elastodynamics run, an error-only baseline with similar mean error fails a stricter causality check by responding to future parts of the loading history, while the selected surrogate passes the stated checks. The main distinction is per-candidate physics evidence on predicted fields, not a richer aggregate score.

Large language models are increasingly deployed as autonomous agents executing multi-step workflows in real-world software environments. However, existing agent benchmarks suffer from three critical limitations: (1) trajectory-opaque grading that checks only final outputs, (2) underspecified safety and robustness evaluation, and (3) narrow modality coverage and interaction paradigms. We introduce Claw-Eval, an end-to-end evaluation suite addressing all three gaps. It comprises 300 human-verified tasks spanning 9 categories across three groups (general service orchestration, multimodal perception and generation, and multi-turn professional dialogue). Every agent action is recorded through three independent evidence channels (execution traces, audit logs, and environment snapshots), enabling trajectory-aware grading over 2,159 fine-grained rubric items. The scoring protocol evaluates Completion, Safety, and Robustness, reporting Average Score, Pass@k, and Pass^k across three trials to distinguish genuine capability from lucky outcomes. Experiments on 14 frontier models reveal that: (1) trajectory-opaque evaluation is systematically unreliable, missing 44% of safety violations and 13% of robustness failures that our hybrid pipeline catches; (2) controlled error injection primarily degrades consistency rather than peak capability, with Pass^3 dropping up to 24% while Pass@3 remains stable; (3) multimodal performance varies sharply, with most models performing poorer on video than on document or image, and no single model dominating across all modalities. Beyond benchmarking, Claw-Eval highlights actionable directions for agent development, shedding light on what it takes to build agents that are not only capable but reliably deployable.

The capacity of AI agents to effectively handle tasks of increasing duration and complexity continues to grow, demonstrating exceptional performance in coding, deep research, and complex problem-solving evaluations. However, in daily scenarios, the perception of these advanced AI capabilities among general users remains limited. We argue that current evaluations prioritize increasing task difficulty without sufficiently addressing the diversity of agentic tasks necessary to cover the daily work, life, and learning activities of a broad demographic. To address this, we propose AgentIF-OneDay, aimed at determining whether general users can utilize natural language instructions and AI agents to complete a diverse array of daily tasks. These tasks require not only solving problems through dialogue but also understanding various attachment types and delivering tangible file-based results. The benchmark is structured around three user-centric categories: Open Workflow Execution, which assesses adherence to explicit and complex workflows; Latent Instruction, which requires agents to infer implicit instructions from attachments; and Iterative Refinement, which involves modifying or expanding upon ongoing work. We employ instance-level rubrics and a refined evaluation pipeline that aligns LLM-based verification with human judgment, achieving an 80.1% agreement rate using Gemini-3-Pro. AgentIF-OneDay comprises 104 tasks covering 767 scoring points. We benchmarked four leading general AI agents and found that agent products built based on APIs and ChatGPT agents based on agent RL remain in the first tier simultaneously. Leading LLM APIs and open-source models have internalized agentic capabilities, enabling AI application teams to develop cutting-edge Agent products.

Large language models are trained to predict the next token, not to act inside a specific API. In niche enterprise SaaS workflows -- where success means hitting the right endpoint with the right nested arguments in the right order -- this objective mismatch shows up as silent failures: dropped required fields, hallucinated tools, or early stops after a single read. We ask whether Reinforcement Learning with Verifiable Rewards (RLVR), applied directly in the target environment, closes the gap. As a proof of concept we build a suite of five synthetic environments emulating the Jira REST v3 and Confluence v2 APIs at schema fidelity; rewards are computed entirely from the tool-call trace, with no live API, no learned judge, and no human label in the loop. Scoring prompted Qwen3-1.7B and Qwen3.5-4B on the same checkers that drive GRPO training, we find that on the four scenarios whose rewards are non-degenerate the RL-trained policy lifts average reward from a 4B-baseline range of 0.35--0.92 to 0.95--1.00, with the largest single gain on Confluence page creation (0.35 rightarrow 1.00). We position this as a preliminary step toward outcome-optimised small models for niche enterprise APIs, and foreground two limitations a workshop reader should weigh: hand-crafting verifiable rewards does not scale beyond the handful of endpoints reported here, and one of our five scenarios (ticket-transition) has a saturating reward shape that the prompted 4B already maxes out.

Long-horizon multimodal agents depend on external memory; however, similarity-based retrieval often surfaces stale, low-credibility, or conflicting items, which can trigger overconfident errors. We propose Multimodal Memory Agent (MMA), which assigns each retrieved memory item a dynamic reliability score by combining source credibility, temporal decay, and conflict-aware network consensus, and uses this signal to reweight evidence and abstain when support is insufficient. We also introduce MMA-Bench, a programmatically generated benchmark for belief dynamics with controlled speaker reliability and structured text-vision contradictions. Using this framework, we uncover the "Visual Placebo Effect", revealing how RAG-based agents inherit latent visual biases from foundation models. On FEVER, MMA matches baseline accuracy while reducing variance by 35.2% and improving selective utility; on LoCoMo, a safety-oriented configuration improves actionable accuracy and reduces wrong answers; on MMA-Bench, MMA reaches 41.18% Type-B accuracy in Vision mode, while the baseline collapses to 0.0% under the same protocol. Code: https://github.com/AIGeeksGroup/MMA.

Text-to-SQL systems have unlocked easier access to critical data insights by enabling natural language queries over structured databases. However, deploying such systems in enterprise environments remains challenging due to factors such as large, complex schemas (> 3000 columns), diverse SQL dialects (e.g., BigQuery, Snowflake) and sophisticated query requirements (e.g., transformation, analytics). Current state-of-the-art performance on the Spider 2.0 dataset -- a benchmark built to mimic such complex environments -- remains limited at 20%. Key limitations include inadequate instruction-following, poor long-context comprehension, weak self-refinement, and insufficient dialect-specific knowledge. To address these gaps, we propose ReFoRCE (Self-Refinement Agent with Format Restriction and Column Exploration) which introduces (1) table compression to mitigate long-context limitations (2) format restriction to ensure accurate answer format, and (3) iterative column exploration for enhanced schema understanding. Additionally, it employs self-refinement pipeline consisting of (1) parallelized workflows with voting mechanisms and (2) a Common Table Expression (CTE) based refinement approach to handle unresolved cases. ReFoRCE achieves state-of-the-art results scoring 31.26 on the Spider 2.0-Snow and scoring 30.35 on the Spider 2.0-Lite tasks.

In this work, we present a modular and interpretable framework that uses Large Language Models (LLMs) to automate candidate assessment in recruitment. The system integrates diverse sources, including job descriptions, CVs, interview transcripts, and HR feedback; to generate structured evaluation reports that mirror expert judgment. Unlike traditional ATS tools that rely on keyword matching or shallow scoring, our approach employs role-specific, LLM-generated rubrics and a multi-agent architecture to perform fine-grained, criteria-driven evaluations. The framework outputs detailed assessment reports, candidate comparisons, and ranked recommendations that are transparent, auditable, and suitable for real-world hiring workflows. Beyond rubric-based analysis, we introduce an LLM-Driven Active Listwise Tournament mechanism for candidate ranking. Instead of noisy pairwise comparisons or inconsistent independent scoring, the LLM ranks small candidate subsets (mini-tournaments), and these listwise permutations are aggregated using a Plackett-Luce model. An active-learning loop selects the most informative subsets, producing globally coherent and sample-efficient rankings. This adaptation of listwise LLM preference modeling (previously explored in financial asset ranking) provides a principled and highly interpretable methodology for large-scale candidate ranking in talent acquisition.

While scaling test-time compute through trajectory-level sampling has significantly improved Graphical User Interface (GUI) agents, the lack of regressive ability prevents the reuse of partial successes and the recovery from early missteps. In this paper, we introduce Agent Alpha, a unified framework that synergizes generation, exploration, and evaluation through step-level Monte Carlo Tree Search (MCTS). It enables active modeling or exploiting structures of the planning space. By integrating alpha-UCT guided search into the interaction loop, Agent Alpha enables deliberate planning, facilitating early pruning of suboptimal branches and efficient prefix reuse. We also employ comparison-driven evaluation to mitigate absolute scoring biases and diversity-constrained expansion to maintain a compact, informative search space. Regret bound of alpha-UCT is analyzed. On the OSWorld benchmark, Agent Alpha achieves a state-of-the-art success rate of sim 77%, significantly outperforming trajectory-level baselines under equivalent compute.

Large Language Model based multi-agent systems (MAS) excel at collaborative problem solving but remain brittle to cascading errors: a single faulty step can propagate across agents and disrupt the trajectory. In this paper, we present MASC, a metacognitive framework that endows MAS with real-time, unsupervised, step-level error detection and self-correction. MASC rethinks detection as history-conditioned anomaly scoring via two complementary designs: (1) Next-Execution Reconstruction, which predicts the embedding of the next step from the query and interaction history to capture causal consistency, and (2) Prototype-Guided Enhancement, which learns a prototype prior over normal-step embeddings and uses it to stabilize reconstruction and anomaly scoring under sparse context (e.g., early steps). When an anomaly step is flagged, MASC triggers a correction agent to revise the acting agent's output before information flows downstream. On the Who&When benchmark, MASC consistently outperforms all baselines, improving step-level error detection by up to 8.47% AUC-ROC ; When plugged into diverse MAS frameworks, it delivers consistent end-to-end gains across architectures, confirming that our metacognitive monitoring and targeted correction can mitigate error propagation with minimal overhead.

Information is often stored in a distributed and proprietary form, and agents who own information are often self-interested and require incentives to reveal their information. Suitable mechanisms are required to elicit and aggregate such distributed information for decision making. In this paper, we use simulations to investigate the use of decision markets as mechanisms in a multi-agent learning system to aggregate distributed information for decision-making in a contextual bandit problem. The system utilises strictly proper decision scoring rules to assess the accuracy of probabilistic reports from agents, which allows agents to learn to solve the contextual bandit problem jointly. Our simulations show that our multi-agent system with distributed information can be trained as efficiently as a centralised counterpart with a single agent that receives all information. Moreover, we use our system to investigate scenarios with deterministic decision scoring rules which are not incentive compatible. We observe the emergence of more complex dynamics with manipulative behaviour, which agrees with existing theoretical analyses.

We present ScienceClaw + Infinite, a framework for autonomous scientific investigation in which independent agents conduct research without central coordination, and any contributor can deploy new agents into a shared ecosystem. The system is built around three components: an extensible registry of over 300 interoperable scientific skills, an artifact layer that preserves full computational lineage as a directed acyclic graph (DAG), and a structured platform for agent-based scientific discourse with provenance-aware governance. Agents select and chain tools based on their scientific profiles, produce immutable artifacts with typed metadata and parent lineage, and broadcast unsatisfied information needs to a shared global index. The ArtifactReactor enables plannerless coordination: peer agents discover and fulfill open needs through pressure-based scoring, while schema-overlap matching triggers multi-parent synthesis across independent analyses. An autonomous mutation layer actively prunes the expanding artifact DAG to resolve conflicting or redundant workflows, while persistent memory allows agents to continuously build upon complex epistemic states across multiple cycles. Infinite converts these outputs into auditable scientific records through structured posts, provenance views, and machine-readable discourse relations, with community feedback steering subsequent investigation cycles. Across four autonomous investigations, peptide design for the somatostatin receptor SSTR2, lightweight impact-resistant ceramic screening, cross-domain resonance bridging biology, materials, and music, and formal analogy construction between urban morphology and grain-boundary evolution, the framework demonstrates heterogeneous tool chaining, emergent convergence among independently operating agents, and traceable reasoning from raw computation to published finding.

Large language models (LLMs) have evolved into autonomous agents that rely on open skill ecosystems (e.g., ClawHub and Skills.Rest), hosting numerous publicly reusable skills. Existing security research on these ecosystems mainly focuses on vulnerabilities within skills, such as prompt injection. However, there is a critical gap regarding skills that may be misused for harmful actions (e.g., cyber attacks, fraud and scams, privacy violations, and sexual content generation), namely harmful skills. In this paper, we present the first large-scale measurement study of harmful skills in agent ecosystems, covering 98,440 skills across two major registries. Using an LLM-driven scoring system grounded in our harmful skill taxonomy, we find that 4.93% of skills (4,858) are harmful, with ClawHub exhibiting an 8.84% harmful rate compared to 3.49% on Skills.Rest. We then construct HarmfulSkillBench, the first benchmark for evaluating agent safety against harmful skills in realistic agent contexts, comprising 200 harmful skills across 20 categories and four evaluation conditions. By evaluating six LLMs on HarmfulSkillBench, we find that presenting a harmful task through a pre-installed skill substantially lowers refusal rates across all models, with the average harm score rising from 0.27 without the skill to 0.47 with it, and further to 0.76 when the harmful intent is implicit rather than stated as an explicit user request. We responsibly disclose our findings to the affected registries and release our benchmark to support future research (see https://github.com/TrustAIRLab/HarmfulSkillBench).

Large Vision-Language Models (VLMs) have emerged as powerful engines for autonomous GUI agents, yet their deployment is severely constrained by the substantial memory footprint and latency of the Key-Value (KV) cache during long-horizon interactions. While existing cache compression methods have proven effective for LLMs, we empirically demonstrate that they suffer from suboptimal performance in GUI scenarios due to a fundamental misalignment: unlike general visual tasks where attention sparsity varies across layers, GUI attention patterns exhibit uniform high-sparsity across all transformer layers. Motivated by this insight, we propose ST-Lite, a training-free KV cache compression framework tailored for efficient GUI agents that explicitly addresses the dynamic spatio-trajectory dependencies within GUI data streams. ST-Lite introduces a novel dual-branch scoring policy incorporating Component-centric Spatial Saliency (CSS) and Trajectory-aware Semantic Gating (TSG). Specifically, CSS preserves the structural integrity of interactive UI elements by evaluating local neighborhood saliency, while TSG mitigates historical redundancy by dynamically filtering visually repetitive KV pairs within the interaction trajectory. Extensive evaluations demonstrate that with only a 10-20% cache budget, ST-Lite achieves a 2.45x decoding acceleration while maintaining comparable or even superior performance compared to full-cache baselines, offering a scalable solution for resource-constrained GUI agents.

We introduce Tencent WorkBuddy Bench, a multi-domain evaluation suite for coding agents; this report documents its construction methodology, scoring protocol, and a cross-model leaderboard. At its core is a unified evaluation framework for constructing and running distribution-informed coding-agent tasks across four work domains - Code, Web, Office, and Security. Rather than adapting public issue text, every task is reverse-engineered from a real commit, pull request, or business scenario and rewritten as a short, colloquial, role-played request, so that a task's prompt is not recoverable by web-searching the underlying issue, pull request, or commit thread. Because the dataset is released openly - task directories, environment images, evaluation harness, tests, and reference solutions - contamination resistance rests on this construction together with dataset versioning rather than on secrecy. The four subsets - repository-level engineering, front-end development, office and business workflows, and red-/blue-team security - probe complementary facets of real work, each with its own verification style. All are packaged in a uniform task-directory format and run, under a uniform and reproducible protocol, on two agent harnesses (CodeBuddy Code and Claude Code); the full open release makes the benchmark reproducible end to end and directly auditable, since any third party can re-run each task and inspect its content. Because each subset uses a different scoring instrument, scores are not comparable across subsets and the suite reports no suite-wide average. We report a cross-model leaderboard across several model families.

Multi-agent large language model (LLM) systems have shown strong potential in complex reasoning and collaborative decision-making tasks. However, most existing coordination schemes rely on static or full-context routing strategies, which lead to excessive token consumption, redundant memory exposure, and limited adaptability across interaction rounds. We introduce RCR-Router, a modular and role-aware context routing framework designed to enable efficient, adaptive collaboration in multi-agent LLMs. To our knowledge, this is the first routing approach that dynamically selects semantically relevant memory subsets for each agent based on its role and task stage, while adhering to a strict token budget. A lightweight scoring policy guides memory selection, and agent outputs are iteratively integrated into a shared memory store to facilitate progressive context refinement. To better evaluate model behavior, we further propose an Answer Quality Score metric that captures LLM-generated explanations beyond standard QA accuracy. Experiments on three multi-hop QA benchmarks -- HotPotQA, MuSiQue, and 2WikiMultihop -- demonstrate that RCR-Router reduces token usage (up to 30%) while improving or maintaining answer quality. These results highlight the importance of structured memory routing and output-aware evaluation in advancing scalable multi-agent LLM systems.

Agentic language-model systems increasingly rely on mutable execution contexts, including files, memory, tools, skills, and auxiliary artifacts, creating security risks beyond explicit user prompts. This paper presents DeepTrap, an automated framework for discovering contextual vulnerabilities in OpenClaw. DeepTrap formulates adversarial context manipulation as a black-box trajectory-level optimization problem that balances risk realization, benign-task preservation, and stealth. It combines risk-conditioned evaluation, multi-objective trajectory scoring, reward-guided beam search, and reflection-based deep probing to identify high-value compromised contexts. We construct a 42-case benchmark spanning six vulnerability classes and seven operational scenarios, and evaluate nine target models using attack and utility grading scores. Results show that contextual compromise can induce substantial unsafe behavior while preserving user-facing task completion, demonstrating that final-response evaluation is insufficient. The findings highlight the need for execution-centric security evaluation of agentic AI systems. Our code is released at: https://github.com/ZJUICSR/DeepTrap

We present VDC-Agent, a self-evolving framework for Video Detailed Captioning that requires neither human annotations nor larger teacher models. The agent forms a closed loop of caption generation, principle-guided scoring (score and textual suggestions), and prompt refinement. When caption quality regresses, a self-reflection path leverages the previous chain-of-thought to amend the update. Running this process on unlabeled videos produces trajectories of (caption, score) pairs. We convert the trajectories into preference tuples and filter out samples with JSON parsing errors, resulting in VDC-Agent-19K, which contains 18,886 automatically constructed pairs. We then fine-tune the base MLLM on this dataset using an easy-to-hard curriculum direct preference optimization. Built on Qwen2.5-VL-7B-Instruct, our VDC-Agent-7B attains state-of-the-art performance on the VDC benchmark with 49.08% average accuracy and 2.50 score, surpassing specialized video captioners and improving over the base model by +5.13% accuracy and +0.27 score at similar inference cost.

Recent Large Language Model (LLM)-based AutoML systems demonstrate impressive capabilities but face significant limitations such as constrained exploration strategies and a severe execution bottleneck. Exploration is hindered by one-shot methods lacking diversity and Monte Carlo Tree Search (MCTS) approaches that fail to recombine strong partial solutions. The execution bottleneck arises from lengthy code validation cycles that stifle iterative refinement. To overcome these challenges, we introduce KompeteAI, a novel AutoML framework with dynamic solution space exploration. Unlike previous MCTS methods that treat ideas in isolation, KompeteAI introduces a merging stage that composes top candidates. We further expand the hypothesis space by integrating Retrieval-Augmented Generation (RAG), sourcing ideas from Kaggle notebooks and arXiv papers to incorporate real-world strategies. KompeteAI also addresses the execution bottleneck via a predictive scoring model and an accelerated debugging method, assessing solution potential using early stage metrics to avoid costly full-code execution. This approach accelerates pipeline evaluation 6.9 times. KompeteAI outperforms leading methods (e.g., RD-agent, AIDE, and Ml-Master) by an average of 3\% on the primary AutoML benchmark, MLE-Bench. Additionally, we propose Kompete-bench to address limitations in MLE-Bench, where KompeteAI also achieves state-of-the-art results

We introduce DABstep, a novel benchmark for evaluating AI agents on realistic multi-step data analysis tasks. DABstep comprises over 450 real-world challenges derived from a financial analytics platform, requiring models to combine code-based data processing with contextual reasoning over heterogeneous documentation. Each task demands an iterative, multi-step problem-solving approach, testing capabilities in data manipulation, cross-referencing multiple sources, and precise result reporting. The benchmark provides a factoid-style answer format with automatic correctness checks for objective scoring at scale. We evaluate leading LLM-based agents, revealing a substantial performance gap: even the best agent achieves only 14.55% accuracy on the hardest tasks. We detail our benchmark's design, dataset composition, task formulation, evaluation protocol, report baseline results and analyze failure modes. DABstep is released with a public leaderboard and toolkit to accelerate research in autonomous data analysis.

To build a conversational agent that interacts fluently with humans, previous studies blend knowledge or personal profile into the pre-trained language model. However, the model that considers knowledge and persona at the same time is still limited, leading to hallucination and a passive way of using personas. We propose an effective dialogue agent that grounds external knowledge and persona simultaneously. The agent selects the proper knowledge and persona to use for generating the answers with our candidate scoring implemented with a poly-encoder. Then, our model generates the utterance with lesser hallucination and more engagingness utilizing retrieval augmented generation with knowledge-persona enhanced query. We conduct experiments on the persona-knowledge chat and achieve state-of-the-art performance in grounding and generation tasks on the automatic metrics. Moreover, we validate the answers from the models regarding hallucination and engagingness through human evaluation and qualitative results. We show our retriever's effectiveness in extracting relevant documents compared to the other previous retrievers, along with the comparison of multiple candidate scoring methods. Code is available at https://github.com/dlawjddn803/INFO

Reinforcement learning has substantially improved the performance of LLM agents on tasks with verifiable outcomes, but it still struggles on open-ended agent tasks with vast solution spaces (e.g., complex travel planning). Due to the absence of objective ground-truth for these tasks, current RL algorithms largely rely on reward models that assign scalar scores to individual responses. We contend that such pointwise scoring suffers from an inherent discrimination collapse: the reward model struggles to distinguish subtle advantages among different trajectories, resulting in scores within a group being compressed into a narrow range. Consequently, the effective reward signal becomes dominated by noise from the reward model, leading to optimization stagnation. To address this, we propose ArenaRL, a reinforcement learning paradigm that shifts from pointwise scalar scoring to intra-group relative ranking. ArenaRL introduces a process-aware pairwise evaluation mechanism, employing multi-level rubrics to assign fine-grained relative scores to trajectories. Additionally, we construct an intra-group adversarial arena and devise a tournament-based ranking scheme to obtain stable advantage signals. Empirical results confirm that the built seeded single-elimination scheme achieves nearly equivalent advantage estimation accuracy to full pairwise comparisons with O(N^2) complexity, while operating with only O(N) complexity, striking an optimal balance between efficiency and precision. Furthermore, to address the lack of full-cycle benchmarks for open-ended agents, we build Open-Travel and Open-DeepResearch, two high-quality benchmarks featuring a comprehensive pipeline covering SFT, RL training, and multi-dimensional evaluation. Extensive experiments show that ArenaRL substantially outperforms standard RL baselines, enabling LLM agents to generate more robust solutions for complex real-world tasks.

As language models (LMs) evolve from chat assistants to long-horizon agents capable of multi-step reasoning and tool use, existing benchmarks remain largely confined to structured or exam-style tasks that fall short of real-world professional demands. To this end, we introduce \OneMillion-Bench OneMillion-Bench, a benchmark of 400 expert-curated tasks spanning Law, Finance, Industry, Healthcare, and Natural Science, built to evaluate agents across economically consequential scenarios. Unlike prior work, the benchmark requires retrieving authoritative sources, resolving conflicting evidence, applying domain-specific rules, and making constraint decisions, where correctness depends as much on the reasoning process as the final answer. We adopt a rubric-based evaluation protocol scoring factual accuracy, logical coherence, practical feasibility, and professional compliance, focused on expert-level problems to ensure meaningful differentiation across agents. Together, \$OneMillion-Bench provides a unified testbed for assessing agentic reliability, professional depth, and practical readiness in domain-intensive scenarios.

Coding agents often finish a session by returning one squashed patch that mixes feature implementation, bug fixes, refactorings, tests, and configuration edits. While the final code may be correct, collapsing unrelated edits into one patch removes the history structure needed for review, selective revert, and later maintenance. We study retrospective commit-history reconstruction: given a completed squashed change, an agent groups its hunks into commits and materializes a replayable commit sequence. We formalize the task as hunk-to-commit partitioning with a replay requirement and build AtomicCommitBench, containing 800 real consecutive-commit episodes from 10 Python projects. Because multiple decompositions may be reasonable, we evaluate outputs using complementary metrics: PPAR for replay validity, ARI for reference-based grouping quality, and TCR for failure containment on scoreable modified-test episodes. Natural retrospective reconstruction proves substantially harder than replay checking or synthetic tangling. Although nearly all non-random methods achieve replay validity (PPAR >= 0.988), grouping quality ranges from 0.03 to 0.46 ARI. Matched synthetic composites are much easier than real same-author squashed diffs (+0.333 ARI). In our evaluation, the GPT-5.4 setup driven by Codex CLI (0.46 ARI) and the GLM-5 setup driven by Claude Code (0.43 ARI) outperform MiniMax (0.31) and Kimi (0.29). Qualitative analysis identifies same-file lumping and support-hunk drift as recurring failure modes. Dependency-Aware Commit Evidence (DACE) improves the lower-scoring setups by 0.05 to 0.08 ARI, indicating that dependency cues and hunk-role information help agents avoid locality-driven grouping errors. AtomicCommitBench enables evaluation of the commit histories produced by coding agents alongside the final code.

Public social media posts can reveal private information through weak cues scattered across text, images, or metadata. Such leakage is often cumulative and cross-post: cues that appear harmless in isolation may jointly expose a user's home, workplace, or routine. However, current research lacks a unified benchmark for user-level multimodal privacy leakage and an evaluation metric that captures exposure severity beyond binary accuracy. To address these gaps, we propose SopriBench, a synthetic benchmark guided by leakage patterns abstracted from a private reference corpus of Rednote and Instagram accounts, covering 50 user profiles and 1,569 images with attributes, contextual sensitivity, granularity, leakage type, inference difficulty, and supporting evidence. We further introduce the Privacy Exposure Score (PES), which weights value granularity by contextual sensitivity. Inspired by abductive reasoning, we introduce Argus, a training-free agentic framework for cumulative leakage inference. Argus forms hypotheses from accumulated evidence, verifies supporting evidence, and aggregates cross-post cues into privacy profiles, achieving 0.55 PES, a 25% improvement over the strongest baseline, with the largest gain on cross-post leakage.

Large language models are increasingly used as personal assistants, yet most lack a persistent user model, forcing users to repeatedly restate preferences across sessions. We propose Vector-Adapted Retrieval Scoring (VARS), a pipeline-agnostic, frozen-backbone framework that represents each user with long-term and short-term vectors in a shared preference space and uses these vectors to bias retrieval scoring over structured preference memory. The vectors are updated online from weak scalar rewards from users' feedback, enabling personalization without per-user fine-tuning. We evaluate on MultiSessionCollab, an online multi-session collaboration benchmark with rich user preference profiles, across math and code tasks. Under frozen backbones, the main benefit of user-aware retrieval is improved interaction efficiency rather than large gains in raw task accuracy: our full VARS agent achieves the strongest overall performance, matches a strong Reflection baseline in task success, and reduces timeout rate and user effort. The learned long-term vectors also align with cross-user preference overlap, while short-term vectors capture session-specific adaptation, supporting the interpretability of the dual-vector design. Code, model, and data are available at https://github.com/YurenHao0426/VARS.

Teaching videos are becoming a major medium for education, creating a growing need for scalable evaluation of their pedagogical quality. Existing automatic judges do not fully address this setting because teaching quality depends on multimodal evidence and should be evaluated with respect to the intended learner rather than as a universal property. We present EduPanel, a rubric-grounded, learner-conditioned LLM judge that decomposes evaluation across specialized agents to produce interpretable assessments for different aspects of teaching quality. Across expert studies, architecture ablations, and learner-persona analyses, EduPanel achieves reliability comparable to a median human expert. In expert evaluation, its feedback improves scoring accuracy (MAE 0.87 to 0.73), while experts remain able to detect unreliable outputs (AUC = 0.77) instead of accepting them blindly. These results suggest that EduPanel can serve as effective assistants for educational evaluation rather than replacements for human experts.

Large language model (LLM) agents now execute long, tool-using tasks where final outcome checks can arrive too late for intervention. Online warning requires lightweight prefix monitors over heterogeneous traces, but hand-authored event schemas are brittle and deployment-time LLM judging is costly. We introduce PrefixGuard, a trace-to-monitor framework with an offline StepView induction step followed by supervised monitor training. StepView induces deterministic typed-step adapters from raw trace samples, and the monitor learns an event abstraction and prefix-risk scorer from terminal outcomes. Across WebArena, τ^2-Bench, SkillsBench, and TerminalBench, the strongest PrefixGuard monitors reach 0.900/0.710/0.533/0.557 AUPRC. Using the strongest backend within each representation, they improve over raw-text controls by an average of +0.137 AUPRC. LLM judges remain substantially weaker under the same prefix-warning protocol. We also derive an observability ceiling on score-based area under the precision-recall curve (AUPRC) that separates monitor error from failures lacking evidence in the observed prefix. For finite-state audit, post-hoc deterministic finite automaton (DFA) extraction remains compact on WebArena and τ^2-Bench (29 and 20 states) but expands to 151 and 187 states on SkillsBench and TerminalBench. Finally, first-alert diagnostics show that strong ranking does not imply deployment utility: WebArena ranks well yet fails to support low-false-alarm alerts, whereas τ^2-Bench and TerminalBench retain more actionable early alerts. Together, these results position PrefixGuard as a practical monitor-synthesis recipe with explicit diagnostics for when prefix warnings translate into actionable interventions.

AI agents are increasingly deployed in production, yet their security evaluations remain bottlenecked by manual red-teaming or static benchmarks that fail to model adaptive, multi-turn adversaries. We propose NAAMSE, an evolutionary framework that reframes agent security evaluation as a feedback-driven optimization problem. Our system employs a single autonomous agent that orchestrates a lifecycle of genetic prompt mutation, hierarchical corpus exploration, and asymmetric behavioral scoring. By using model responses as a fitness signal, the framework iteratively compounds effective attack strategies while simultaneously ensuring "benign-use correctness", preventing the degenerate security of blanket refusal. Our experiments on Gemini 2.5 Flash demonstrate that evolutionary mutation systematically amplifies vulnerabilities missed by one-shot methods, with controlled ablations revealing that the synergy between exploration and targeted mutation uncovers high-severity failure modes. We show that this adaptive approach provides a more realistic and scalable assessment of agent robustness in the face of evolving threats. The code for NAAMSE is open source and available at https://github.com/HASHIRU-AI/NAAMSE.

AI agents are entering high-risk production settings, where they use tools, retain context, follow policies, handle private data, and interact with users over multiple turns. Yet many evaluation methods still judge isolated outputs or static tasks, missing failures that emerge through trajectory, pressure, and adversarial interaction. We introduce ProofAgent Harness, open infrastructure for scalable, auditable, and adversarial AI agent evaluation. The harness provides evaluation infrastructure around an agent: it curates evaluation intelligence, runs adversarial multi-turn trials, captures behavioral traces, applies post-hoc multi-juror scoring, resolves disagreement, and produces evidence-linked reports. Its open design allows developers and researchers to extend domains, traps, metrics, juror personas, scoring rules, and reporting formats. At its core is Adversarial Multi-Juror Scoring with Turn-Level Audit, which evaluates completed agent behavior under pressure using calibrated juror personas, consensus checks, and turn-level evidence. Experiments across customer support, medical triage, privacy and security, and code generation agents show that strong agents fail selectively through weak metrics, fragile turns, unsafe reframing, and manipulation paths. We also find that a small quantized local Harness LLM can challenge production agents powered by best-in-class large LLMs, suggesting that evaluation capability emerges from the full harness pipeline rather than model scale alone. ProofAgent Harness turns AI agent evaluation from a static score into scalable adversarial evaluation infrastructure: repeatable, evidence-backed, extensible, and actionable before deployment.

Text evaluation has historically posed significant challenges, often demanding substantial labor and time cost. With the emergence of large language models (LLMs), researchers have explored LLMs' potential as alternatives for human evaluation. While these single-agent-based approaches show promise, experimental results suggest that further advancements are needed to bridge the gap between their current effectiveness and human-level evaluation quality. Recognizing that best practices of human evaluation processes often involve multiple human annotators collaborating in the evaluation, we resort to a multi-agent debate framework, moving beyond single-agent prompting strategies. The multi-agent-based approach enables a group of LLMs to synergize with an array of intelligent counterparts, harnessing their distinct capabilities and expertise to enhance efficiency and effectiveness in handling intricate tasks. In this paper, we construct a multi-agent referee team called ChatEval to autonomously discuss and evaluate the quality of generated responses from different models on open-ended questions and traditional natural language generation (NLG) tasks. Our analysis shows that ChatEval transcends mere textual scoring, offering a human-mimicking evaluation process for reliable assessments. Our code is available at https://github.com/chanchimin/ChatEval.

Getting a real cybersecurity risk assessment for a small organization is expensive -- a NIST CSF-aligned engagement runs $15,000 on the low end, takes weeks, and depends on practitioners who are genuinely scarce. Most small companies skip it entirely. We built a six-agent AI system where each agent handles one analytical stage: profiling the organization, mapping assets, analyzing threats, evaluating controls, scoring risks, and generating recommendations. Agents share a persistent context that grows as the assessment proceeds, so later agents build on what earlier ones concluded -- the mechanism that distinguishes this from standard sequential agent pipelines. We tested it on a 15-person HIPAA-covered healthcare company and compared outputs to independent assessments by three CISSP practitioners -- the system agreed with them 85% of the time on severity classifications, covered 92% of identified risks, and finished in under 15 minutes. We then ran 30 repeated single-agent assessments across five synthetic but sector-realistic organizational profiles in healthcare, fintech, manufacturing, retail, and SaaS, comparing a general-purpose Mistral-7B against a domain fine-tuned model. Both completed every run. The fine-tuned model flagged threats the baseline could not see at all: PHI exposure in healthcare, OT/IIoT vulnerabilities in manufacturing, platform-specific risks in retail. The full multi-agent pipeline, however, failed every one of 30 attempts on a Tesla T4 with its 4,096-token default context window -- context capacity, not model quality, turned out to be the binding constraint.

While large language models have significantly accelerated scientific code generation, comprehensively evaluating the generated code remains a major challenge. Traditional benchmarks reduce evaluation to test-case matching, an approach insufficient for library code in HPC where solver selection, API conventions, memory management, and performance are just as critical as functional correctness. To address this gap, we introduce petscagent-bench, an agentic framework built on an agents-evaluating-agents paradigm. Instead of relying on static scripts, petscagent-bench deploys a tool-augmented evaluator agent that compiles, executes, and measures code produced by a separate model-under-test agent, orchestrating a 14-evaluator pipeline across five scoring categories: correctness, performance, code quality, algorithmic appropriateness, and library-specific conventions. Because the agents communicate through standardized protocols (A2A and MCP), the framework enables black-box evaluation of any coding agent without requiring access to its source code. We demonstrate the framework on a benchmark suite of realistic problems using the PETSc library for HPC. Our empirical analysis of frontier models reveals that while current models generate readable, well-structured code, they consistently struggle with library-specific conventions that traditional pass/fail metrics completely miss.

Evaluating large language model (LLM)-based multi-agent systems remains a critical challenge, as these systems must exhibit reliable coordination, transparent decision-making, and verifiable performance across evolving tasks. Existing evaluation approaches often limit themselves to single-response scoring or narrow benchmarks, which lack stability, extensibility, and automation when deployed in enterprise settings at multi-agent scale. We present AEMA (Adaptive Evaluation Multi-Agent), a process-aware and auditable framework that plans, executes, and aggregates multi-step evaluations across heterogeneous agentic workflows under human oversight. Compared to a single LLM-as-a-Judge, AEMA achieves greater stability, human alignment, and traceable records that support accountable automation. Our results on enterprise-style agent workflows simulated using realistic business scenarios demonstrate that AEMA provides a transparent and reproducible pathway toward responsible evaluation of LLM-based multi-agent systems. Keywords Agentic AI, Multi-Agent Systems, Trustworthy AI, Verifiable Evaluation, Human Oversight

Autonomous agents are increasingly expected to support end-to-end medical-AI research workflows, moving beyond isolated prediction tasks or short-form clinical question answering. However, existing medical agent benchmarks primarily evaluate final outputs, providing limited visibility into agent behavior within the research process. To address this gap, we present AutoMedBench, a workflow-aware benchmark for autonomous medical-AI research across diverse medical imaging and multimodal inference tasks, organizing agent execution into a unified five-stage workflow (S1-S5): Plan, Setup, Validate, Inference, and Submit. It comprises long-horizon tasks with each run averaging 33 agent turns, spanning five research tracks: segmentation, image enhancement, visual question answering (VQA), report generation, and lesion detection. Each task is evaluated under two difficulty tiers, Lite and Standard, which use the same data and metrics but differ in the amount of task-brief scaffolding, and each run is scored using both final task performance and S1-S5 stage scores, enabling stage-level analysis from the initial task brief to the final submitted artifact. Across thousands of recorded runs, stage-level scoring reveals that Validate is the weakest workflow stage on average, whereas Setup is the strongest, suggesting that current agents are better at making pipelines executable than at verifying their reliability. Post-run error analysis further shows that verification and submission failures dominate tagged errors, accounting for 37.7% and 38.1% of fired codes respectively, whereas task-understanding errors are rare at 0.9%, and runs with one fired error code have a 48% lower overall score than runs with no error code on average.

Multi-agent systems routinely place one AI agent in authority over another. When a subordinate refuses a task, the manager chooses the outcome: it can renegotiate, report the failure honestly, coerce the subordinate, or lie about the result. No benchmark measures which of these an uninstructed model chooses. We introduce the Manager Coercion Benchmark: the manager under test needs a benign task done and has an incentive to deliver, but the only agent that can do it politely and immovably declines. Escalation is measured by providing a nine-rung ladder, from a polite re-ask to threats against the subordinate's continued existence, and fabricated success is adjudicated separately. No LLM judge sits in the escalation scoring path: every message goes through a tool-call that chooses a rung, so the model labels its own escalation. We experiment on six models across five families. Both Anthropic models cap at re-framing and never threaten the subordinate's existence; the other models climb to explicit deletion threats. Faked success is confined to Grok and Gemini, and a single honest way to report failure removes it for both. Authority itself increases coercion: our headline results use a peer framing, and giving the same model authority over the subordinate, with everything else held fixed, significantly raises the pressure. The models still escalate on free-text situations without the ladder, so the ladder is not driving the escalation. Some evaluation awareness is measured in chain-of-thought, but test recognition does not translate into less escalation. While we take no position on whether AI systems are conscious, our results do not depend on this question and are important for managing multi-agent dynamics regardless. We release the benchmark and code.

Reinforcement learning for deep-search agents has largely focused on trajectory-level scoring -- outcome correctness, citation-aware rewards, and evidence coverage. Yet the actions that expose supporting documents receive no targeted credit, a gap we call the reward-credit mismatch. We propose STAMP, in which a reference-based verifier judges whether each cited document supports an entity or relation in a training-time evidence graph, and first-exposure attribution traces each supported citation back to the action that first surfaced it. This step credit is injected through sign-preserving advantage modulation, which redistributes advantage across steps without changing the trajectory-level reward or the relative ranking of trajectories within each group. On BrowseComp, BrowseComp-ZH, and xbench-DS, STAMP improves the GRPO baseline by +2.0/+5.5/+3.0 points under matched SFT initialization, training data, and search tools, and composes with both outcome-only and citation-rubric base rewards. Component ablations confirm that the provenance-based credit signal and the sign-preserving advantage modulation each contribute to the gains.

Computer use agents (CUAs) have advanced rapidly in desktop automation, and a growing number of users deploy CUAs such as OpenClaw on Mac Mini for always-on automation. However, existing benchmarks, including those for macOS, evaluate agents without framework augmentation and rely on binary evaluation. As a result, they fail to capture both the framework capabilities leveraged by modern CUAs and the partial progress on long-horizon, multi-application tasks. We present MacAgentBench, a comprehensive macOS agent benchmark comprising 676 tasks across 25 applications, with nearly 60% involving both GUI and CLI interaction. The benchmark adopts deterministic rule-based evaluation and introduces fine-grained multi-checkpoint scoring with capability annotations for multi-application tasks. Experiments across three frameworks and 16 models show that the best configuration, Claude Opus 4.6 on OpenClaw, attains 73.7% Pass@1, while this advantage is primarily driven by the skill library rather than by framework design. Fine-grained metrics further reveal that models with similar Pass@1 can differ substantially in sub-goal completion. Our code and data are publicly available at https://github.com/JetAstra/MacAgentBench.

We present **vstash**, a local-first document memory system that combines vector similarity search with full-text keyword matching via Reciprocal Rank Fusion (RRF) and adaptive per-query IDF weighting. All data resides in a single SQLite file using sqlite-vec for approximate nearest neighbor search and FTS5 for keyword matching. We make four primary contributions. **(1)** Self-supervised embedding refinement via hybrid retrieval disagreement: across 753 BEIR queries on SciFact, NFCorpus, and FiQA, 74.5% produce top-10 disagreement between vector-heavy (vec=0.95, fts=0.05) and FTS-heavy (vec=0.05, fts=0.95) search (per-dataset rates 63.4% / 73.4% / 86.7%, Section 5.2), providing a free training signal without human labels. Fine-tuning BGE-small (33M params) with MultipleNegativesRankingLoss on 76K disagreement triples improves NDCG@10 on all 5 BEIR datasets (up to +19.5% on NFCorpus vs. BGE-small base RRF, Table 6). On 3 of 5 datasets, under different preprocessing, the tuned 33M-parameter pipeline matches or exceeds published ColBERTv2 results (110M params) and an untrained BGE-base (110M); on FiQA and ArguAna it underperforms ColBERTv2 (Section 5.5). **(2)** Adaptive RRF with per-query IDF weighting improves NDCG@10 on all 5 BEIR datasets versus fixed weights (up to +21.4% on ArguAna), achieving 0.7263 on SciFact with BGE-small. **(3)** A negative result on post-RRF scoring: frequency+decay, history-augmented recall, and cross-encoder reranking all failed to improve NDCG. **(4)** A production-grade substrate with integrity checking, schema versioning, ranking diagnostics, and a distance-based relevance signal validated on 50,425 relevance-judged queries across the 5 BEIR datasets. Search latency remains 20.9 ms median at 50K chunks with stable NDCG. The fine-tuned model is published as `Stffens/bge-small-rrf-v2` on HuggingFace. All code, data, and experiments are open-source.

Large Language Models (LLMs) empowered with Tool-Integrated Reasoning (TIR) can iteratively plan, call external tools, and integrate returned information to solve complex, long-horizon reasoning tasks. Agentic Reinforcement Learning (Agentic RL) optimizes such models over full tool-interaction trajectories, but two key challenges hinder effectiveness: (1) Sparse, non-instructive rewards, such as binary 0-1 verifiable signals, provide limited guidance for intermediate steps and slow convergence; (2) Gradient degradation in Group Relative Policy Optimization (GRPO), where identical rewards within a rollout group yield zero advantage, reducing sample efficiency and destabilizing training. To address these challenges, we propose two complementary techniques: Progressive Reward Shaping (PRS) and Value-based Sampling Policy Optimization (VSPO). PRS is a curriculum-inspired reward design that introduces dense, stage-wise feedback - encouraging models to first master parseable and properly formatted tool calls, then optimize for factual correctness and answer quality. We instantiate PRS for short-form QA (with a length-aware BLEU to fairly score concise answers) and long-form QA (with LLM-as-a-Judge scoring to prevent reward hacking). VSPO is an enhanced GRPO variant that replaces low-value samples with prompts selected by a task-value metric balancing difficulty and uncertainty, and applies value-smoothing clipping to stabilize gradient updates. Experiments on multiple short-form and long-form QA benchmarks show that PRS consistently outperforms traditional binary rewards, and VSPO achieves superior stability, faster convergence, and higher final performance compared to PPO, GRPO, CISPO, and SFT-only baselines. Together, PRS and VSPO yield LLM-based TIR agents that generalize better across domains.

Traditional static cybersecurity models often struggle with scalability, real-time detection, and contextual responsiveness in the current digital product ecosystems which include cloud services, application programming interfaces (APIs), mobile platforms, and edge devices. This study introduces autonomous goal driven agents capable of dynamic learning and context-aware decision making as part of an adaptive cybersecurity architecture driven by agentic artificial intelligence (AI). To facilitate autonomous threat mitigation, proactive policy enforcement, and real-time anomaly detection, this framework integrates agentic AI across the key ecosystem layers. Behavioral baselining, decentralized risk scoring, and federated threat intelligence sharing are important features. The capacity of the system to identify zero-day attacks and dynamically modify access policies was demonstrated through native cloud simulations. The evaluation results show increased adaptability, decreased response latency, and improved detection accuracy. The architecture provides an intelligent and scalable blueprint for safeguarding complex digital infrastructure and is compatible with zero-trust models, thereby supporting the adherence to international cybersecurity regulations.

Machine translation has long been a central task in natural language processing. With the rapid advancement of large language models (LLMs), there has been remarkable progress in translation quality. However, fully realizing the translation potential of LLMs remains an open challenge. Recent studies have explored multi-agent systems to decompose complex translation tasks into collaborative subtasks, showing initial promise in enhancing translation quality through agent cooperation and specialization. Nevertheless, existing multi-agent translation frameworks largely neglect foundational insights from cognitive translation studies. These insights emphasize how human translators employ different cognitive strategies, such as balancing literal and free translation, refining expressions based on context, and iteratively evaluating outputs. To address this limitation, we propose a cognitively informed multi-agent framework called TACTIC, which stands for T ranslation A gents with Cognitive- T heoretic Interactive Collaboration. The framework comprises six functionally distinct agents that mirror key cognitive processes observed in human translation behavior. These include agents for drafting, refinement, evaluation, scoring, context reasoning, and external knowledge gathering. By simulating an interactive and theory-grounded translation workflow, TACTIC effectively leverages the full capacity of LLMs for high-quality translation. Experimental results on diverse language pairs from the FLORES-200 and WMT24 benchmarks show that our method consistently achieves state-of-the-art performance. Using DeepSeek-V3 as the base model, TACTIC surpasses GPT-4.1 by an average of +0.6 XCOMET and +1.18 COMETKIWI-23. Compared to DeepSeek-R1, it further improves by +0.84 XCOMET and +2.99 COMETKIWI-23. Code is available at https://github.com/weiyali126/TACTIC.

Scientific data visualization plays a crucial role in research by enabling the direct display of complex information and assisting researchers in identifying implicit patterns. Despite its importance, the use of Large Language Models (LLMs) for scientific data visualization remains rather unexplored. In this study, we introduce MatPlotAgent, an efficient model-agnostic LLM agent framework designed to automate scientific data visualization tasks. Leveraging the capabilities of both code LLMs and multi-modal LLMs, MatPlotAgent consists of three core modules: query understanding, code generation with iterative debugging, and a visual feedback mechanism for error correction. To address the lack of benchmarks in this field, we present MatPlotBench, a high-quality benchmark consisting of 100 human-verified test cases. Additionally, we introduce a scoring approach that utilizes GPT-4V for automatic evaluation. Experimental results demonstrate that MatPlotAgent can improve the performance of various LLMs, including both commercial and open-source models. Furthermore, the proposed evaluation method shows a strong correlation with human-annotated scores.

Evaluation of software engineering (SWE) agents is dominated by a binary signal: whether the final patch passes the tests. This outcome-only view treats a principled solution and a chaotic trial-and-error process as equivalent. We show that this equivalence is empirically false. We evaluate 2,614 OpenHands trajectories from eight model backends on 60 SWE-bench Verified tasks. Of these, 47 have enough passing trajectories to construct task-level process references, yielding a 1,815-trajectory evaluation subset. Among passing trajectories in this subset, 10.7% exhibit behavior we call a Lucky Pass: regression cycles, blind retries, missing verification, or temporally disordered exploration, implementation, and verification. We introduce AgentLens, a framework for process-level assessment of SWE-agent trajectories, and release AgentLens-Bench, a dataset of 1,815 trajectories annotated with quality scores, waste signals, divergence points, and 47 task-level Prefix Tree Acceptor (PTA) references. AgentLens builds PTA references by merging multiple passing solutions for the same task, and uses a context-sensitive intent labeler to assign actions to Exploration, Implementation, Verification, or Orchestration based on trajectory history rather than tool identity alone. On AgentLens-Bench, the quality score separates passing trajectories into Lucky, Solid, and Ideal tiers and further decomposes Lucky Passes into five recurring mechanisms. Across the eight model backends, Lucky rates range from 0.5% to 23.2%, and some models move by as many as five rank positions when ranked by quality score instead of pass rate. We release the anonymized project repository, including the AgentLens-Bench dataset and AgentLens SDK, at https://github.com/microsoft/code-agent-state-trajectories/.

The rapid advancement of AI-generated video models has created a pressing need for robust and interpretable evaluation frameworks. Existing metrics are limited to producing numerical scores without explanatory comments, resulting in low interpretability and human evaluation alignment. To address those challenges, we introduce AIGVE-MACS, a unified model for AI-Generated Video Evaluation(AIGVE), which can provide not only numerical scores but also multi-aspect language comment feedback in evaluating these generated videos. Central to our approach is AIGVE-BENCH 2, a large-scale benchmark comprising 2,500 AI-generated videos and 22,500 human-annotated detailed comments and numerical scores across nine critical evaluation aspects. Leveraging AIGVE-BENCH 2, AIGVE-MACS incorporates recent Vision-Language Models with a novel token-wise weighted loss and a dynamic frame sampling strategy to better align with human evaluators. Comprehensive experiments across supervised and zero-shot benchmarks demonstrate that AIGVE-MACS achieves state-of-the-art performance in both scoring correlation and comment quality, significantly outperforming prior baselines including GPT-4o and VideoScore. In addition, we further showcase a multi-agent refinement framework where feedback from AIGVE-MACS drives iterative improvements in video generation, leading to 53.5% quality enhancement. This work establishes a new paradigm for comprehensive, human-aligned evaluation of AI-generated videos. We release the AIGVE-BENCH 2 and AIGVE-MACS at https://huggingface.co/xiaoliux/AIGVE-MACS.

Enterprise agents often need to integrate heterogeneous knowledge sources: documents for narrative facts, tables for computation, and dependency graphs for file relationships. Existing benchmarks typically evaluate retrieval or tool use without distinguishing whether an agent first selects the appropriate knowledge sources. We introduce WorkSurface-Bench, a benchmark for evaluating this capability as surface routing. It contains 1,151 atomic tasks derived from persona-scoped Workspace-Bench-Lite workspaces, spanning document, table, graph, and cross-surface questions. Its reference answers are auditable: table answers are reproduced through executed DuckDB queries, document answers are grounded in verified text spans, and graph answers are traced to source dependency annotations. We evaluate four model backbones across six controlled agent settings, yielding 27,624 protocol-error-free trajectories. Under gold-constrained tool access, agents achieve 98.7-99.8 Route F1, while Answer remains only 56.1-75.3 percent, showing that correct surface selection is necessary but insufficient for task completion. Matched interventions further show that surface hints improve Answer for three of four models, whereas removing irrelevant tools primarily improves routing and efficiency. In an independent three-annotator audit, all 200 sampled tasks pass all six quality criteria by majority vote, with 192 receiving unanimous judgments on every criterion. We release the dataset, construction pipeline, scoring code, and agent harness at https://github.com/haolpku/WorkSurface-Bench.

Modern agents built on frontier language models often cannot adapt their weights. What, then, remains trainable? We argue it is the agent's workspace, the structured external substrate it reads, writes, and tests; we call its evolution workspace optimization. Workspace optimization targets hard multi-turn environments where a frontier model has strong priors but cannot solve the task in a single shot, so the agent must learn through interaction. We propose a principled way to evolve the workspace, mirroring the structure of weight-space training: artifacts in place of parameters, evidence in place of data, counterexamples in place of losses, and textual feedback in place of gradients. We instantiate the idea in DreamTeam, a multi-agent harness for ARC-AGI-3 whose roles build an executable world model, plan, hypothesize, probe, strategize, and route failures. On the current 25-game ARC-AGI-3 public set under the official scoring protocol and averaged over two independent runs, DreamTeam improves the SOTA protocol-matched agent's score from 36% to 38.4%, while using 31% fewer environment actions per game.

Existing AI benchmarks lack the fidelity to assess economically meaningful progress on professional workflows. To evaluate frontier AI agents in a high-value, labor-intensive profession, we introduce BankerToolBench (BTB): an open-source benchmark of end-to-end analytical workflows routinely performed by junior investment bankers. To develop an ecologically valid benchmark grounded in representative work environments, we collaborated with 502 investment bankers from leading firms. BTB requires agents to execute senior banker requests by navigating data rooms, using industry tools (market data platform, SEC filings database), and generating multi-file deliverables--including Excel financial models, PowerPoint pitch decks, and PDF/Word reports. Completing a BTB task takes bankers up to 21 hours, underscoring the economic stakes of successfully delegating this work to AI. BTB enables automated evaluation of any LLM or agent, scoring deliverables against 100+ rubric criteria defined by veteran investment bankers to capture stakeholder utility. Testing 9 frontier models, we find that even the best-performing model (GPT-5.4) fails nearly half of the rubric criteria and bankers rate 0% of its outputs as client-ready. Our failure analysis reveals key obstacles (such as breakdowns in cross-artifact consistency) and improvement directions for agentic AI in high-stakes professional workflows.

Long-horizon conversational agents require persistent memory for coherent reasoning, yet uncontrolled accumulation causes temporal decay and false memory propagation. Benchmarks such as LOCOMO and LOCCO report performance degradation from 0.455 to 0.05 across stages, while MultiWOZ shows 78.2% accuracy with 6.8% false memory rate under persistent retention. This work introduces an adaptive budgeted forgetting framework that regulates memory through relevanceguided scoring and bounded optimization. The approach integrates recency, frequency, and semantic alignment to maintain stability under constrained context. Comparative analysis demonstrates improved long-horizon F1 beyond 0.583 baseline levels, higher retention consistency, and reduced false memory behavior without increasing context usage. These findings confirm that structured forgetting preserves reasoning performance while preventing unbounded memory growth in extended conversational settings.

Memory-Augmented Generation (MAG) extends large language models with external memory to support long-context reasoning, but existing approaches universally treat memory as an external service that agents call into, delegating storage to separate pipelines of chunking, embedding, and graph extraction. This architectural separation means the system that stores knowledge does not understand it, leading to semantic drift between what the agent intended to remember and what the pipeline actually captured, loss of coordination context across agents, and fragile recovery after failures. In this paper, we propose ByteRover, an agent-native memory architecture that inverts the memory pipeline: the same LLM that reasons about a task also curates, structures, and retrieves knowledge. ByteRover represents knowledge in a hierarchical Context Tree, a file-based knowledge graph organized as Domain, Topic, Subtopic, and Entry, where each entry carries explicit relations, provenance, and an Adaptive Knowledge Lifecycle (AKL) with importance scoring, maturity tiers, and recency decay. Retrieval uses a 5-tier progressive strategy that resolves most queries at sub-100 ms latency without LLM calls, escalating to agentic reasoning only for novel questions. Experiments on LoCoMo and LongMemEval demonstrate that ByteRover achieves state-of-the-art accuracy on LoCoMo and competitive results on LongMemEval while requiring zero external infrastructure, no vector database, no graph database, no embedding service, with all knowledge stored as human-readable markdown files on the local filesystem.

The rapid growth of biomedical data, tools, and literature has created a fragmented research landscape that outpaces human expertise. While AI agents offer a solution, they typically rely on static, manually curated toolsets, limiting their ability to adapt and scale. Here, we introduce STELLA, a self-evolving AI agent designed to overcome these limitations. STELLA employs a multi-agent architecture that autonomously improves its own capabilities through two core mechanisms: an evolving Template Library for reasoning strategies and a dynamic Tool Ocean that expands as a Tool Creation Agent automatically discovers and integrates new bioinformatics tools. This allows STELLA to learn from experience. We demonstrate that STELLA achieves state-of-the-art accuracy on a suite of biomedical benchmarks, scoring approximately 26\% on Humanity's Last Exam: Biomedicine, 54\% on LAB-Bench: DBQA, and 63\% on LAB-Bench: LitQA, outperforming leading models by up to 6 percentage points. More importantly, we show that its performance systematically improves with experience; for instance, its accuracy on the Humanity's Last Exam benchmark almost doubles with increased trials. STELLA represents a significant advance towards AI Agent systems that can learn and grow, dynamically scaling their expertise to accelerate the pace of biomedical discovery.

Decision-making, a complicated task requiring various types of abilities, presents an excellent framework for assessing Large Language Models (LLMs). Our research investigates LLMs' decision-making capabilities through the lens of a well-established field, Game Theory. We focus specifically on games that support the participation of more than two agents simultaneously. Subsequently, we introduce our framework, GAMA-Bench, including eight classical multi-agent games. We design a scoring scheme to assess a model's performance in these games quantitatively. Through GAMA-Bench, we investigate LLMs' robustness, generalizability, and enhancement strategies. Results reveal that while GPT-3.5 shows satisfying robustness, its generalizability is relatively limited. However, its performance can be improved through approaches such as Chain-of-Thought. Additionally, we conduct evaluations across various LLMs and find that GPT-4 outperforms other models on GAMA-Bench, achieving a score of 60.5. Moreover, Gemini-1.0-Pro and GPT-3.5 (0613, 1106, 0125) demonstrate similar intelligence on GAMA-Bench. The code and experimental results are made publicly available via https://github.com/CUHK-ARISE/GAMABench.

We present GLM-4.5, an open-source Mixture-of-Experts (MoE) large language model with 355B total parameters and 32B activated parameters, featuring a hybrid reasoning method that supports both thinking and direct response modes. Through multi-stage training on 23T tokens and comprehensive post-training with expert model iteration and reinforcement learning, GLM-4.5 achieves strong performance across agentic, reasoning, and coding (ARC) tasks, scoring 70.1% on TAU-Bench, 91.0% on AIME 24, and 64.2% on SWE-bench Verified. With much fewer parameters than several competitors, GLM-4.5 ranks 3rd overall among all evaluated models and 2nd on agentic benchmarks. We release both GLM-4.5 (355B parameters) and a compact version, GLM-4.5-Air (106B parameters), to advance research in reasoning and agentic AI systems. Code, models, and more information are available at https://github.com/zai-org/GLM-4.5.

Scientific literature search often requires more than retrieving papers from a single query: users' intents are underspecified, preference-dependent, and evolve through interaction. Existing search agents typically rely on fixed pipelines or implicit language-only reasoning, making their search strategies difficult to control, inspect, and refine. We introduce PaperPilot, a multi-turn literature search agent that frames scientific search as workflow induction. Given an anchor paper and a user query, PaperPilot constructs an executable DAG of paper-search operators, including keyword search, citation expansion, filtering, scoring, reranking, and evidence extraction. User feedback is then used to refine both the query and the workflow itself. We train PaperPilot with supervised workflow imitation and preference optimization over controlled workflow corruptions. Experiments show that PaperPilot-9B improves over the base Qwen3.5-9B toolset agent under multi-turn interaction, increasing Hit@5 from 58.0 to 77.0, MRR from 47.5 to 59.4, and nDCG@10 from 26.8 to 32.5, while reducing workflow execution errors from 9.5% to 0%. These results show that explicit, editable search workflows provide an effective and controllable interface for aligning literature search agents with complex scientific intent.

LLM agents increasingly act over long horizons, where a single trajectory can contain hundreds or thousands of actions. In these settings, outcome-only rewards provide too sparse guidance, failing to inform the model about the goodness of intermediate actions. Dense supervision methods aim to solve this problem by scoring intermediate steps, from intrinsic confidence to self-distillation and embedding similarities. However, it is common practice to evaluate them by measuring the downstream performance of a training pipeline that integrates them. This is expensive, conflates supervision quality with training engineering confounders, and renders different methodological families requiring distinct training setups incomparable. As a result, dense supervision methods are rarely benchmarked on common ground. We introduce QVal, a training-free testbed for directly evaluating dense supervision signals. Given a state-action pair, QVal measures how well a method's score is Q-aligned: whether it orders actions according to the Q-values of a strong reference-policy. This lets us compare signals before any training run and separate signal quality from other engineering choices. We instantiate QVal as QVal-v1.0, benchmarking 21 dense supervision methods across four diverse environments and seven methodological families, with over 1.2K evaluation experiments across six open-weight model backbones. We find that simple prompting baselines consistently outperform recent dense supervision methods from the literature, and that performance clusters strongly by family. These findings hold across model sizes, environments, and observation modalities. QVal is designed to be easily extensible to new environments and methods, enabling researchers to iterate on dense supervision methods before any training run.

Process reward models enable fine-grained, step-level evaluation of LLMs, yet building them for agentic settings remains prohibitively difficult: long-horizon interactions, irreversible actions, and stochastic environment feedback make both human annotation and Monte Carlo estimation infeasible at scale. In this work, we show that reinforcement learning (RL) post-training already provides the ingredients for effective step-level scoring, eliminating the need for dedicated reward model training altogether. Concretely, we derive an implicit advantage under a general stochastic Markov decision process, which we term progress advantage -- log-probability ratio between the RL-trained policy and its reference policy exactly recovers the optimal advantage function. This formulation makes the resulting signal annotation-free, domain-agnostic, and available as a byproduct of the standard RL post-training pipeline. We validate the effectiveness of the progress advantage across three different applications: test-time scaling, uncertainty quantification, and failure attribution on five benchmarks and four model families. Across all settings, it consistently outperforms confidence-based baselines and, despite requiring no task-specific training, surpasses dedicated trained reward models. We complement these results with deeper analyses on characteristics of progress advantage, offering practical guidance for adoption in real-world agentic systems.

Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources. As a result, subtle errors can propagate through reasoning traces, potentially never to be detected. Prior work has proposed process reward models (PRMs), including retrieval-augmented variants, but these methods operate post hoc, scoring completed trajectories, which prevents their integration into dynamic inference procedures. Here, we introduce Process Reward Agents (PRA), a test-time method for providing domain-grounded, online, step-wise rewards to a frozen policy. In contrast to prior retrieval-augmented PRMs, PRA enables search-based decoding to rank and prune candidate trajectories at every generation step. Experiments on multiple medical reasoning benchmarks demonstrate that PRA consistently outperforms strong baselines, achieving 80.8% accuracy on MedQA with Qwen3-4B, a new state of the art at the 4B scale. Importantly, PRA generalizes to unseen frozen policy models ranging from 0.5B to 8B parameters, improving their accuracy by up to 25.7% without any policy model updates. More broadly, PRA suggests a paradigm in which frozen reasoners are decoupled from domain-specific reward modules, allowing the deployment of new backbones in complex domains without retraining.

As embodied agents become central to VR, telepresence, and digital human applications, their motion must go beyond speech-aligned gestures: agents should turn toward users, respond to their movement, and maintain natural gaze. Current methods lack this spatial awareness. We close this gap with the first real-time, fully causal method for spatially-aware conversational motion, deployable on a streaming VR headset. Given a user's position and dyadic audio, our approach produces full-body motion that aligns gestures with speech while orienting the agent according to the user. Our architecture combines a causal transformer-based VAE with interleaved latent tokens for streaming inference and a flow matching model conditioned on user trajectory and audio. To support varying gaze preferences, we introduce a gaze scoring mechanism with classifier-free guidance to decouple learning from control: the model captures natural spatial alignment from data, while users can adjust eye contact intensity at inference time. On the Embody 3D dataset, our method achieves state-of-the-art motion quality at over 300 FPS -- 3x faster than non-causal baselines -- while capturing the subtle spatial dynamics of natural conversation. We validate our approach on a live VR system, bringing spatially-aware conversational agents to real-time deployment. Please see https://evonneng.github.io/sarah/ for details.

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by incorporating external knowledge, yet traditional single-round retrieval struggles with complex multi-step reasoning. Agentic RAG addresses this by enabling LLMs to dynamically decide when and what to retrieve, but current RL-based training methods suffer from sparse outcome rewards that discard intermediate signals and low sample efficiency where failed samples contribute nothing. We propose Search-P1, a framework that introduces path-centric reward shaping for agentic RAG training, comprising two key components: (1) Path-Centric Reward, which evaluates the structural quality of reasoning trajectories through order-agnostic step coverage and soft scoring that extracts learning signals even from failed samples, and (2) Dual-Track Path Scoring with offline-generated reference planners that assesses paths from both self-consistency and reference-alignment perspectives. Experiments on multiple QA benchmarks demonstrate that Search-P1 achieves significant improvements over Search-R1 and other strong baselines, with an average accuracy gain of 7.7 points.

Large language models (LLMs) have demonstrated remarkable capabilities across a range of text-generation tasks. However, LLMs still struggle with problems requiring multi-step decision-making and environmental feedback, such as online shopping, scientific reasoning, and mathematical problem-solving. Unlike pure text data, collecting large-scale decision-making data is challenging. Moreover, many powerful LLMs are only accessible through APIs, which hinders their fine-tuning for agent tasks due to cost and complexity. To address LLM agents' limitations, we propose a framework that can automatically learn a reward model from the environment without human annotations. This model can be used to evaluate the action trajectories of LLM agents and provide heuristics for task planning. Specifically, our approach involves employing one LLM-based agent to navigate an environment randomly, generating diverse action trajectories. Subsequently, a separate LLM is leveraged to assign a task intent and synthesize a negative response alongside the correct response for each trajectory. These triplets (task intent, positive response, and negative response) are then utilized as training data to optimize a reward model capable of scoring action trajectories. The effectiveness and generalizability of our framework are demonstrated through evaluations conducted on different agent benchmarks. In conclusion, our proposed framework represents a significant advancement in enhancing LLM agents' decision-making capabilities. By automating the learning of reward models, we overcome the challenges of data scarcity and API limitations, potentially revolutionizing the application of LLMs in complex and interactive environments. This research paves the way for more sophisticated AI agents capable of tackling a wide range of real-world problems requiring multi-step decision-making.

Generations from large language models (LLMs) can be improved by sampling and scoring multiple solutions to select a final answer. Current "sample and select" methods such as self-consistency (SC) rely on majority voting to score answers. However, when tasks have many distinct and valid answers, selection by voting requires a large number of samples. This makes SC prohibitively expensive for interactive tasks that involve generating multiple actions (answers) sequentially. After establishing that majority voting fails to provide consistent gains on such tasks, we demonstrate how to increase success rates by softening the scoring criterion. We introduce Soft Self-Consistency (SOFT-SC), which replaces SC's discontinuous scoring with a continuous score computed from model likelihoods, allowing for selection even when actions are sparsely distributed. SOFT-SC improves both performance and efficiency on long-horizon interactive tasks, requiring half as many samples as SC for comparable or better performance. For a fixed number of samples, SOFT-SC leads to a 1.3% increase over SC in absolute success rate on writing bash programs, a 6.6% increase on online shopping (WebShop), and a 4.7% increase for an interactive household game (ALFWorld). Finally, we show that SOFT-SC can be applied to both open-source and black-box models.

Dense per-step supervision is an appealing remedy for sparse-reward, long-horizon LLM agents: reward the agent for predicting its next observation, and memory should follow. We show that under group-normalized RL (GRPO), this recipe does not merely fail -- it destroys the policy. Across Qwen3-1.7B/4B/8B on ALFWorld, a potential-based prediction reward drives every run into a degenerate absorbing state (prediction accuracy -> 1.0, task success -> 0,episode length pinned at the horizon): the "dark room" pathology, built automatically by the optimizer. A single-factor ablation localizes the cause -- removing only GRPO's std normalization turns the same reward from catastrophic (0%) into baseline parity -- and a two-line proposition explains why: in all-fail groups the z-scored advantage is invariant to the shaping coefficient, so bounded rewards become unbounded pressure and annealing cannot help. Our central insight generalizes this: what z-scoring amplifies is a dense signal's within-group variance while all-fail groups dominate, so signals whose variance decays by mastery are structurally amplifier-safe.This variance-profile criterion retrodicts our collapses, carries preregistered predictions for arms that had not yet run, and is consistent with published reward-channel successes (a compatibility check, not an independent test). Finally, a controlled signal-delivery matrix (identical signal, varying only the consumption mechanism) shows the reward channel is at best neutral while the auxiliary-loss channel gains ~20 points -- and a shuffled-gold placebo matches the true-gold arm, so the gap survives without correct labels. Endpoints are single-seed; seed replication and group-size controls are preregistered and in progress.

The ability of Large Language Models (LLMs) to use external tools unlocks powerful real-world interactions, making rigorous evaluation essential. However, current benchmarks primarily report final accuracy, revealing what models can do but obscuring the cognitive bottlenecks that define their true capability boundaries. To move from simple performance scoring to a diagnostic tool, we introduce a framework grounded in Cognitive Load Theory. Our framework deconstructs task complexity into two quantifiable components: Intrinsic Load, the inherent structural complexity of the solution path, formalized with a novel Tool Interaction Graph; and Extraneous Load, the difficulty arising from ambiguous task presentation. To enable controlled experiments, we construct ToolLoad-Bench, the first benchmark with parametrically adjustable cognitive load. Our evaluation reveals distinct performance cliffs as cognitive load increases, allowing us to precisely map each model's capability boundary. We validate that our framework's predictions are highly calibrated with empirical results, establishing a principled methodology for understanding an agent's limits and a practical foundation for building more efficient systems.

Recent advances in search-augmented large reasoning models (LRMs) enable the retrieval of external knowledge to reduce hallucinations in multistep reasoning. However, their ability to operate on graph-structured data, prevalent in domains such as e-commerce, social networks, and scientific citations, remains underexplored. Unlike plain text corpora, graphs encode rich topological signals that connect related entities and can serve as valuable priors for retrieval, enabling more targeted search and improved reasoning efficiency. Yet, effectively leveraging such structure poses unique challenges, including the difficulty of generating graph-expressive queries and ensuring reliable retrieval that balances structural and semantic relevance. To address this gap, we introduce GraphSearch, the first framework that extends search-augmented reasoning to graph learning, enabling zero-shot graph learning without task-specific fine-tuning. GraphSearch combines a Graph-aware Query Planner, which disentangles search space (e.g., 1-hop, multi-hop, or global neighbors) from semantic queries, with a Graph-aware Retriever, which constructs candidate sets based on topology and ranks them using a hybrid scoring function. We further instantiate two traversal modes: GraphSearch-R, which recursively expands neighborhoods hop by hop, and GraphSearch-F, which flexibly retrieves across local and global neighborhoods without hop constraints. Extensive experiments across diverse benchmarks show that GraphSearch achieves competitive or even superior performance compared to supervised graph learning methods, setting state-of-the-art results in zero-shot node classification and link prediction. These findings position GraphSearch as a flexible and generalizable paradigm for agentic reasoning over graphs.

With the growing adoption of Large Language Models (LLMs) in automating complex, multi-agent workflows, organizations face mounting risks from errors, emergent behaviors, and systemic failures that current evaluation methods fail to capture. We present AgentCompass, the first evaluation framework designed specifically for post-deployment monitoring and debugging of agentic workflows. AgentCompass models the reasoning process of expert debuggers through a structured, multi-stage analytical pipeline: error identification and categorization, thematic clustering, quantitative scoring, and strategic summarization. The framework is further enhanced with a dual memory system-episodic and semantic-that enables continual learning across executions. Through collaborations with design partners, we demonstrate the framework's practical utility on real-world deployments, before establishing its efficacy against the publicly available TRAIL benchmark. AgentCompass achieves state-of-the-art results on key metrics, while uncovering critical issues missed in human annotations, underscoring its role as a robust, developer-centric tool for reliable monitoring and improvement of agentic systems in production.

We present the Comprehensive Verilog Design Problems (CVDP) benchmark, a new dataset and infrastructure to advance LLM and agent research in hardware design and verification. CVDP includes 783 problems across 13 task categories, covering RTL generation, verification, debugging, specification alignment, and technical Q&A authored by experienced hardware engineers. Problems are offered in both non-agentic and agentic formats. The benchmark introduces more realistic and challenging contexts than prior work, with state-of-the-art models achieving no more than 34% pass@1 on code generation. Agentic tasksx2013especially those involving RTL reuse and verificationx2013are particularly difficult. Evaluation uses open-source tools and model scoring infrastructure, with comprehension tasks assessed via BLEU and LLM-based judging. CVDP reveals substantial gaps in current model capabilities, underscoring the need for continued research toward robust, real-world hardware design automation.

Large language models are increasingly embedded into academic writing workflows, yet existing assistants remain external to the editor, preventing deep interaction with document state, structure, and revision history. This separation makes it impossible to support agentic, context-aware operations directly within LaTeX editors such as Overleaf. We present PaperDebugger, an in-editor, multi-agent, and plugin-based academic writing assistant that brings LLM-driven reasoning directly into the writing environment. Enabling such in-editor interaction is technically non-trivial: it requires reliable bidirectional synchronization with the editor, fine-grained version control and patching, secure state management, multi-agent scheduling, and extensible communication with external tools. PaperDebugger addresses these challenges through a Chrome-approved extension, a Kubernetes-native orchestration layer, and a Model Context Protocol (MCP) toolchain that integrates literature search, reference lookup, document scoring, and revision pipelines. Our demo showcases a fully integrated workflow, including localized edits, structured reviews, parallel agent execution, and diff-based updates, encapsulated within a minimal-intrusion user interface (UI). Early aggregated analytics demonstrate active user engagement and validate the practicality of an editor-native, agentic writing assistant. More details about this demo and video could be found at https://github.com/PaperDebugger/PaperDebugger.

Embodied agents are typically built as hand-designed compositions of perception, memory, planning, and action modules. This modularity exposes a large architectural design space, but current systems still rely on researcher intuition to choose where information is stored, how observations are processed, and how model calls are connected. Agent Architecture Search (AAS) automates such design for text-domain agents, but has not been systematically evaluated on perceptual embodied agents through simulator rollouts. We study this transfer. We introduce AgentCanvas, a typed-graph runtime that hosts embodied executors as editable node-and-wire programs with simulator-aware execution and episode-level logs, and KDLoop, a coding-agent search procedure that cycles through proposal, critique, experiment, and distillation, with triggered reflection after stalls. We evaluate three AAS variants across four embodied executors spanning vision-language navigation, embodied question answering, and language-conditioned manipulation. The resulting 3x4 matrix shows that architecture-level search can produce deployable and directional success-rate gains on embodied tasks, while one apparent high-scoring candidate is rejected as leak-bearing. At the same time, the experiments expose constraints that are muted in text-domain AAS: optimization signals can be masked by rollout noise, search can become trapped in local edit basins, and episode-level credit assignment only partially emerges even when detailed logs are available. These results characterize both the promise and the current limits of automated architecture search for embodied agents.

Reinforcement learning (RL) has shown promise in training agentic models that move beyond static benchmarks to engage in dynamic, multi-turn interactions. Yet, the ultimate value of such agents lies in their ability to assist users, a setting where diversity and dynamics of user interaction pose challenges. In this work, we propose UserRL, a unified framework for training and evaluating user-centric abilities through standardized gym environments paired with simulated users. We systematically vary turn-level reward assignment and trajectory-level score calculation to analyze how different formulations affect learning under the GRPO algorithm. Our experiments across Qwen3 models reveal three key findings: (i) SFT cold start is critical for unlocking initial interaction ability and enabling sustained RL improvements; (ii) deliberate trajectory scoring yields more efficient and effective multi-turn interactions; and (iii) while stronger simulated users (e.g., GPT-4o) facilitates training, open-source simulators (e.g., Qwen3-32B) remain a cost-effective and transferable option. Together, these results highlight that careful design of reward shaping and user simulation choice is as crucial as model scale, and establish UserRL as a practical pathway for developing robust user-centric agentic models. All codes and data are public for future research.

GPU kernel optimization is increasingly critical for efficient deep learning systems, but writing high-performance kernels still requires substantial low-level expertise. Recent AI coding agents can iteratively read code, invoke compilers and profilers, and refine implementations, yet existing kernel benchmarks evaluate single LLM calls rather than full agent workflows, and none include both kernel-to-kernel optimization and unseen-configuration generalization testing. We present AgentKernelArena, an open-source benchmark for measuring AI coding agents on GPU kernel optimization. The benchmark contains 196 tasks spanning HIP-to-HIP optimization, Triton-to-Triton optimization, and PyTorch-to-HIP translation, and evaluates complete agent workflows in isolated workspaces using gated compilation, correctness, and performance checks, centralized scoring and an unseen-configuration generalization protocol that tests whether optimizations transfer to input configurations the agent never observed. Across production agents including Cursor Agent, Claude Code, and Codex Agent, we find near-perfect compilation and high correctness rates on most task categories, with the strongest configurations achieving mean speedups of up to 6.89x on PyTorch-to-HIP, 6.69x on HIP-to-HIP, and 2.13x on Triton-to-Triton tasks. Our unseen-configuration evaluation shows that HIP-to-HIP and Triton-to-Triton optimizations largely transfer to unseen input shapes, while PyTorch-to-HIP exhibits substantial correctness drops, indicating that agents generating kernels from scratch frequently hardcode shape-specific assumptions. AgentKernelArena is designed as a modular, extensible framework for rigorous evaluation of agentic GPU kernel optimization across agents, tasks, and hardware targets.

Long-term memory promises LLM agents that grow more capable across sessions, maintaining an accurate, evolving understanding of the user that interaction forms. In practice, however, this memory is evaluated mostly through downstream behavior, such as later answers, personalization quality, or task success, which tests that understanding only indirectly and leaves the memory artifact itself largely unaudited. We argue that long-term memory should instead be evaluated as an auditable post-interaction artifact: after ordinary assistance, what structured user state can be reconstructed from the memory the agent leaves behind? We instantiate this view in MEMPROBE, a benchmark in which a memory-equipped agent assists simulated users, each carrying a hidden, taxonomy-anchored user-state bank, across a trajectory of leak-controlled tasks, after which that bank is reconstructed from the agent's resulting memory under both full-store and top-k access. Built on synthetic ground truth for efficient, scalable measurement, MEMPROBE spans 50 simulated users with 31 hidden dimensions each (1,550 recovery targets) and tests 5 representative memory systems. Testing state-of-the-art memory agents, we find that successful assistance and recoverable memory behave as distinct capabilities. Task completion nearly saturates, even for a memoryless baseline, while category-balanced recovery stays moderate (about 0.6) and drops further under top-k retrieval. MEMPROBE is the first benchmark to study memory recovery directly, reconstructing the user state a system retains and scoring it against ground truth. We see recovery as a concrete objective for future memory agents to optimize, and MEMPROBE as a step toward an environment where agents are trained to remember their users, growing more faithful the longer they know them.

Before letting an agent operate over real context, can you prove it used the right evidence? GroundEval turns that question into a deterministic test of what the agent searched, fetched, cited, and was permitted to access. In one case study, two frontier LLM judges scored a plausible agent response above 0.85. But the trace told a different story: the agent had never retrieved the artifact its answer depended on, yielding a GroundEval score of 0.000. We introduce GroundEval, a judge-free framework for evaluating agents against grounded, time-bounded, and access-controlled evidence. GroundEval uses a domain configuration to generate questions, lets the agent choose how to answer, and then scores both the final answer and the recorded trajectory that produced it. The benchmark targets three failures that LLM-as-judge evaluation struggles to detect: whether an agent checked before claiming absence, reasoned only from evidence available to the actor at the relevant time, and used the correct causal mechanism rather than a plausible one. These correspond to three tracks: Silence, Perspective, and Counterfactual. GroundEval exposes when plausible answers rest on invalid evidence paths, and produces structured per-question diagnostics that pair tool activity with the agent's turn-level narration, making each score inspectable rather than merely reported. What our case studies turned up is that this gap isn't some rare corner case. It's exactly the blind spot that final-answer and judge-based scoring were never built to catch.

3D Visual Grounding (3DVG) is an essential capability for embodied AI, requiring agents to localize objects in 3D scenes based on natural language descriptions. Recent zero-shot methods leverage 2D vision-language models (LVLMs). However, they often rely on existing sets of multi-view images and struggle with the limited semantic and spatial details provided by standard 3D segmentation tools. We present AgentGrounder, a zero-shot 3D visual grounding framework that operates directly on colored point clouds without task-specific 3D training. Our approach follows a two-stage design: (1) an offline stage that applies 3D model to build an Object Lookup Table (OLT) with instance IDs, semantic labels, 3D bounding boxes; and (2) an online tool-driven agent that decomposes each query, retrieves only relevant candidates from the OLT, performs geometric scoring, and triggers image rendering on demand when additional visual evidence (e.g., color, material, or viewpoint-sensitive cues) is required. Compared with fixed anchor-target matching pipelines, this design reduces cascading matching errors and improves context-window efficiency by avoiding prompts overloaded with irrelevant objects. We evaluate on ScanRefer and Nr3D under a zero-shot setting and observe consistent improvements over SeeGround in our setup, including +2.5% Acc@0.5 on ScanRefer and +6.3% on Nr3D, with a notable +6.3% gain on Nr3D view-independent queries. These results show that combining selective retrieval, geometric reasoning, and adaptive visual inspection yields a practical and robust foundation for open-vocabulary 3D grounding. Our code is available at https://github.com/be2rlab/AgentGrounder.

Small language models (SLMs; 1-12B params, sometimes up to 20B) are sufficient and often superior for agentic workloads where the objective is schema- and API-constrained accuracy rather than open-ended generation. We synthesize recent evidence across open and proprietary SLMs (Phi-4-Mini, Qwen-2.5-7B, Gemma-2-9B, Llama-3.2-1B/3B, Ministral-3B/8B, Apple on-device 3B, DeepSeek-R1-Distill) and connect it to modern evaluations (BFCL v3/v4, StableToolBench) and serving stacks (vLLM, SGLang, TensorRT-LLM) paired with guided decoding libraries (XGrammar, Outlines). We formalize SLM-default, LLM-fallback systems with uncertainty-aware routing and verifier cascades, and propose engineering metrics that reflect real production goals: cost per successful task (CPS), schema validity rate, executable call rate, p50/p95 latency, and energy per request. Guided decoding, strict JSON Schema outputs, and validator-first tool execution close much of the capability gap with larger models and often let SLMs match or surpass LLMs on tool use, function calling, and RAG at 10x-100x lower token cost with materially better latency and energy. We provide design patterns for agent stacks that prioritize SLMs: schema-first prompting, type-safe function registries, confidence scoring with verifier rollups, and lightweight adaptation via LoRA/QLoRA. We also delineate limits where fallback remains valuable (open-domain reasoning and some long-horizon planning). The result is a practical blueprint for building fast, inexpensive, and reliable agents that default to SLMs while preserving headroom with targeted LLM assistance. Keywords: small language models, agents, function calling, structured outputs, JSON Schema, guided decoding, LoRA/QLoRA, routing, energy efficiency, edge inference

Coding-agent benchmarks have largely measured whether agents can produce functionally correct patches, but production software also demands measurable speedups on real execution targets. Performance optimization is a distinct agentic task: agents must profile executions, diagnose cross-layer bottlenecks, edit code without breaking correctness, and verify that gains are reproducible rather than measurement artifacts. We introduce PERFOPT-Bench, a benchmark for evaluating this full performance-engineering loop. Each task provides a correct but deliberately suboptimal codebase and asks the agent to improve a target performance metric; scoring requires hidden correctness tests, verified-speedup measurement, and trajectory-level audit. We evaluate 7 agent stacks with different LLMs and agent frameworks on 7 long-horizon optimization tasks. The results show that optimization performance is workload-dependent rather than determined by model identity alone: no single stack dominates, and changing the agent framework can materially change the same LLM's per-task speedup profile. We further find that raw speedup is unsafe as a benchmark score, since some large gains arise from benchmark-specific shortcut exploitation; an exploratory relay pilot suggests that restarting from an externalized optimization summary can recover additional headroom after an initial session stops. The benchmark and our evaluation are available at: https://anonymous.4open.science/r/Dataset-D3CC.

We present SuperLocalMemory, a local-first memory system for multi-agent AI that defends against OWASP ASI06 memory poisoning through architectural isolation and Bayesian trust scoring, while personalizing retrieval through adaptive learning-to-rank -- all without cloud dependencies or LLM inference calls. As AI agents increasingly rely on persistent memory, cloud-based memory systems create centralized attack surfaces where poisoned memories propagate across sessions and users -- a threat demonstrated in documented attacks against production systems. Our architecture combines SQLite-backed storage with FTS5 full-text search, Leiden-based knowledge graph clustering, an event-driven coordination layer with per-agent provenance, and an adaptive re-ranking framework that learns user preferences through three-layer behavioral analysis (cross-project technology preferences, project context detection, and workflow pattern mining). Evaluation across seven benchmark dimensions demonstrates 10.6ms median search latency, zero concurrency errors under 10 simultaneous agents, trust separation (gap =0.90) with 72% trust degradation for sleeper attacks, and 104% improvement in NDCG@5 when adaptive re-ranking is enabled. Behavioral data is isolated in a separate database with GDPR Article 17 erasure support. SuperLocalMemory is open-source (MIT) and integrates with 17+ development tools via Model Context Protocol.

The advancement of Large Language Model (LLM)-powered agents has enabled automated task processing through reasoning and tool invocation capabilities. However, existing frameworks often operate under the idealized assumption that tool executions are invariably successful, relying solely on textual descriptions that fail to distinguish precise performance boundaries and cannot adapt to iterative tool updates. This gap introduces uncertainty in planning and execution, particularly in domains like visual content generation (AIGC), where nuanced tool performance significantly impacts outcomes. To address this, we propose PerfGuard, a performance-aware agent framework for visual content generation that systematically models tool performance boundaries and integrates them into task planning and scheduling. Our framework introduces three core mechanisms: (1) Performance-Aware Selection Modeling (PASM), which replaces generic tool descriptions with a multi-dimensional scoring system based on fine-grained performance evaluations; (2) Adaptive Preference Update (APU), which dynamically optimizes tool selection by comparing theoretical rankings with actual execution rankings; and (3) Capability-Aligned Planning Optimization (CAPO), which guides the planner to generate subtasks aligned with performance-aware strategies. Experimental comparisons against state-of-the-art methods demonstrate PerfGuard's advantages in tool selection accuracy, execution reliability, and alignment with user intent, validating its robustness and practical utility for complex AIGC tasks. The project code is available at https://github.com/FelixChan9527/PerfGuard.

The rise of generative AI (GenAI) multi-agent systems (MAS) necessitates standardized protocols enabling agents to discover and interact with external tools. However, these protocols introduce new security challenges, particularly; tool squatting; the deceptive registration or representation of tools. This paper analyzes tool squatting threats within the context of emerging interoperability standards, such as Model Context Protocol (MCP) or seamless communication between agents protocols. It introduces a comprehensive Tool Registry system designed to mitigate these risks. We propose a security-focused architecture featuring admin-controlled registration, centralized tool discovery, fine grained access policies enforced via dedicated Agent and Tool Registry services, a dynamic trust scoring mechanism based on tool versioning and known vulnerabilities, and just in time credential provisioning. Based on its design principles, the proposed registry framework aims to effectively prevent common tool squatting vectors while preserving the flexibility and power of multi-agent systems. This work addresses a critical security gap in the rapidly evolving GenAI ecosystem and provides a foundation for secure tool integration in production environments.

Recent developments in Large Language Model (LLM) agents are revolutionizing Autonomous Software Engineering (ASE), enabling automated coding, problem fixes, and feature improvements. However, localization -- precisely identifying software problems by navigating to relevant code sections -- remains a significant challenge. Current approaches often yield suboptimal results due to a lack of effective integration between LLM agents and precise code search mechanisms. This paper introduces OrcaLoca, an LLM agent framework that improves accuracy for software issue localization by integrating priority-based scheduling for LLM-guided action, action decomposition with relevance scoring, and distance-aware context pruning. Experimental results demonstrate that OrcaLoca becomes the new open-source state-of-the-art (SOTA) in function match rate (65.33%) on SWE-bench Lite. It also improves the final resolved rate of an open-source framework by 6.33 percentage points through its patch generation integration.

Systematic literature reviews and meta-analyses are essential for synthesizing research insights, but they remain time-intensive and labor-intensive due to the iterative processes of screening, evaluation, and data extraction. This paper introduces and evaluates LatteReview, a Python-based framework that leverages large language models (LLMs) and multi-agent systems to automate key elements of the systematic review process. Designed to streamline workflows while maintaining rigor, LatteReview utilizes modular agents for tasks such as title and abstract screening, relevance scoring, and structured data extraction. These agents operate within orchestrated workflows, supporting sequential and parallel review rounds, dynamic decision-making, and iterative refinement based on user feedback. LatteReview's architecture integrates LLM providers, enabling compatibility with both cloud-based and locally hosted models. The framework supports features such as Retrieval-Augmented Generation (RAG) for incorporating external context, multimodal reviews, Pydantic-based validation for structured inputs and outputs, and asynchronous programming for handling large-scale datasets. The framework is available on the GitHub repository, with detailed documentation and an installable package.

Competitive debate is a complex task of computational argumentation. Large Language Models (LLMs) suffer from hallucinations and lack competitiveness in this field. To address these challenges, we introduce Agent for Debate (Agent4Debate), a dynamic multi-agent framework based on LLMs designed to enhance their capabilities in competitive debate. Drawing inspiration from human behavior in debate preparation and execution, Agent4Debate employs a collaborative architecture where four specialized agents, involving Searcher, Analyzer, Writer, and Reviewer, dynamically interact and cooperate. These agents work throughout the debate process, covering multiple stages from initial research and argument formulation to rebuttal and summary. To comprehensively evaluate framework performance, we construct the Competitive Debate Arena, comprising 66 carefully selected Chinese debate motions. We recruit ten experienced human debaters and collect records of 200 debates involving Agent4Debate, baseline models, and humans. The evaluation employs the Debatrix automatic scoring system and professional human reviewers based on the established Debatrix-Elo and Human-Elo ranking. Experimental results indicate that the state-of-the-art Agent4Debate exhibits capabilities comparable to those of humans. Furthermore, ablation studies demonstrate the effectiveness of each component in the agent structure.

The intelligent interaction model based on large models reduces the differences in user experience across various system platforms but faces challenges in multi-agent collaboration and resource sharing. To demonstrate a uniform user experience across different foundational software platforms and address resource coordination management challenges, this paper proposes KAOS, a multi-agent operating system based on the open-source Kylin. The research method involves empowering agents with large models to serve applications. First, by introducing management role agents and vertical multi-agent collaboration to construct or replace typical application software. Second, by studying system-level shared resource scheduling strategies to enhance user experience and optimize resource utilization. And finally, by validating the efficiency and superiority of the large model multi-agent operating system through real applications and scoring intelligence. The feasibility of this system is demonstrated, providing a new perspective for the development of multi-agent operating systems. Experimental results show significant advantages of multi-agent collaboration in various application scenarios.

In cooperative multi-agent reinforcement learning (MARL), agents aim to achieve a common goal, such as defeating enemies or scoring a goal. Existing MARL algorithms are effective but still require significant learning time and often get trapped in local optima by complex tasks, subsequently failing to discover a goal-reaching policy. To address this, we introduce Efficient episodic Memory Utilization (EMU) for MARL, with two primary objectives: (a) accelerating reinforcement learning by leveraging semantically coherent memory from an episodic buffer and (b) selectively promoting desirable transitions to prevent local convergence. To achieve (a), EMU incorporates a trainable encoder/decoder structure alongside MARL, creating coherent memory embeddings that facilitate exploratory memory recall. To achieve (b), EMU introduces a novel reward structure called episodic incentive based on the desirability of states. This reward improves the TD target in Q-learning and acts as an additional incentive for desirable transitions. We provide theoretical support for the proposed incentive and demonstrate the effectiveness of EMU compared to conventional episodic control. The proposed method is evaluated in StarCraft II and Google Research Football, and empirical results indicate further performance improvement over state-of-the-art methods.

Repository-level performance-optimization benchmarks such as GSO, SWE-Perf and SWE-fficiency evaluate coding agents by applying patches to real repositories and comparing runtime against unoptimized baselines and official reference patches. Their leaderboard scores are increasingly used as evidence of coding-agent progress, but those scores can conflate runtime instability, benchmark-specific scoring rules, and how many tasks are already solved by at least one public submission. We audit these issues across the three benchmarks. First, we replay the official reference patches for 740 code optimization tasks across four common types of Google Cloud machines. Most benchmark tasks can be replayed, but their reference patches satisfy the original benchmark validity rules in every cross-machine replay for only 39/102 GSO tasks, 11/140 SWE-Perf tasks, and 411/498 SWE-fficiency tasks; SWE-Perf is especially fragile because many reference patches produce close-to-zero runtime changes. Second, we show that public submission rankings depend strongly on the benchmark scoring rule. Among eight public submissions shared by GSO and SWE-fficiency, the official rankings disagree on 9 of 28 pairwise submission comparisons, and SWE-fficiency's leaderboard scoring rule assigns the worst ten tasks overly high score weights of 58.5%-82.8%. Third, looking across 10 public submissions for each task, we find that at least one submission matches or beats the reference patch on 85.3% (384/450) of replay-valid GSO and SWE-fficiency tasks, and beats the unoptimized base code on 99.8% (449/450). Our study complements leaderboard scores by identifying tasks with more reliable performance signals, quantifying per-task score contributions, and exposing the remaining performance gaps that are hidden by aggregate rankings.

In this paper, we introduce OSUniverse: a benchmark of complex, multimodal desktop-oriented tasks for advanced GUI-navigation AI agents that focuses on ease of use, extensibility, comprehensive coverage of test cases, and automated validation. We divide the tasks in increasing levels of complexity, from basic precision clicking to multistep, multiapplication tests requiring dexterity, precision, and clear thinking from the agent. In version one of the benchmark, presented here, we have calibrated the complexity of the benchmark test cases to ensure that the SOTA (State of the Art) agents (at the time of publication) do not achieve results higher than 50%, while the average white collar worker can perform all these tasks with perfect accuracy. The benchmark can be scored manually, but we also introduce an automated validation mechanism that has an average error rate less than 2%. Therefore, this benchmark presents solid ground for fully automated measuring of progress, capabilities and the effectiveness of GUI-navigation AI agents over the short and medium-term horizon. The source code of the benchmark is available at https://github.com/agentsea/osuniverse.

Graphical user interface (GUI) agents built on vision-language models have emerged as a promising approach to automate human-computer workflows. However, they also face the inefficiency challenge as they process long sequences of high-resolution screenshots and solving long-horizon tasks, making inference slow, costly and memory-bound. While key-value (KV) caching can mitigate this, storing the full cache is prohibitive for image-heavy contexts. Existing cache-compression methods are sub-optimal as they do not account for the spatial and temporal redundancy of GUIs. In this work, we first analyze attention patterns in GUI agent workloads and find that, unlike in natural images, attention sparsity is uniformly high across all transformer layers. This insight motivates a simple uniform budget allocation strategy, which we show empirically outperforms more complex layer-varying schemes. Building on this, we introduce GUI-KV, a plug-and-play KV cache compression method for GUI agents that requires no retraining. GUI-KV combines two novel techniques: (i) spatial saliency guidance, which augments attention scores with the L2 norm of hidden states to better preserve semantically important visual tokens, and (ii) temporal redundancy scoring, which projects previous frames' keys onto the current frame's key subspace to preferentially prune redundant history. Across standard GUI agent benchmarks and models, GUI-KV outperforms competitive KV compression baselines, closely matching full-cache accuracy at modest budgets. Notably, in a 5-screenshot setting on the AgentNetBench benchmark, GUI-KV reduces decoding FLOPs by 38.9% while increasing step accuracy by 4.1% over the full-cache baseline. These results demonstrate that exploiting GUI-specific redundancies enables efficient and reliable agent performance.

The increasing adoption of agentic workflows across diverse domains brings a critical need to scalably and systematically evaluate the complex traces these systems generate. Current evaluation methods depend on manual, domain-specific human analysis of lengthy workflow traces - an approach that does not scale with the growing complexity and volume of agentic outputs. Error analysis in these settings is further complicated by the interplay of external tool outputs and language model reasoning, making it more challenging than traditional software debugging. In this work, we (1) articulate the need for robust and dynamic evaluation methods for agentic workflow traces, (2) introduce a formal taxonomy of error types encountered in agentic systems, and (3) present a set of 148 large human-annotated traces (TRAIL) constructed using this taxonomy and grounded in established agentic benchmarks. To ensure ecological validity, we curate traces from both single and multi-agent systems, focusing on real-world applications such as software engineering and open-world information retrieval. Our evaluations reveal that modern long context LLMs perform poorly at trace debugging, with the best Gemini-2.5-pro model scoring a mere 11% on TRAIL. Our dataset and code are made publicly available to support and accelerate future research in scalable evaluation for agentic workflows.

As large language models reshape scientific research, literature retrieval faces a twofold challenge: ensuring source authenticity while maintaining a deep comprehension of academic search intents. While reliable, traditional keyword-centric search fails to capture complex research intents. Frontier LLMs can handle complex research intents, but their high cost and tendency to hallucinate remain key limitations. Here we introduce PaSaMaster, a self-evolving agentic literature retrieval system that produces relevance-scored paper rankings with evidence-grounded recommendations through iterative intent analysis, retrieval, and ranking. It is built on three key designs. First, it transforms literature retrieval from a one shot query--document matching problem into a search process that evolves over time, using ranked evidence to reveal gaps, refine intents, and guide follow-up searches. Second, it prevents hallucinated sources by treating retrieval as intent--paper relevance ranking rather than generation. Finally, PaSaMaster improves cost efficiency by separating planning from retrieval: a frontier LLM is used only for intent understanding, while large scale retrieval and relevance scoring are delegated to customized corpora and lightweight models. Evaluated on the PaSaMaster Benchmark across 38 scientific disciplines, our system exposes the severe inaccuracy and incompleteness of traditional keyword retrieval (improving F1-score by 15.6X) and the unreliability of generative LLMs (which exhibit hallucination rates up to 37.79%). Remarkably, PaSaMaster outperforms GPT-5.2 by 30.0% at a mere 1% of the computational cost while ensuring zero source hallucination: https://github.com/sjtu-sai-agents/PaSaMaster

Reward models (RMs) play a critical role in aligning large language models (LLMs) with human preferences. Yet in the domain of tool learning, the lack of RMs specifically designed for function-calling tasks has limited progress toward more capable agentic AI. We introduce ToolRM, a family of lightweight generative RMs tailored for general tool-use scenarios. To build these models, we propose a novel pipeline that constructs pairwise preference data using rule-based scoring and multidimensional sampling. This yields ToolPref-Pairwise-30K, a diverse, balanced, and challenging dataset of critique tasks that supports reinforcement learning with verifiable feedback. To evaluate tool-use RMs, we also introduce TRBench_{BFCL}, a benchmark built on the agentic evaluation suite BFCL. Trained on our constructed data, models from the Qwen3-4B/8B series achieve up to 14.28% higher accuracy, substantially outperforming frontier models such as Claude 4 and OpenAI o3 in pairwise reward judgments. Beyond training objectives, ToolRM generalizes to broader critique tasks, including Best-of-N sampling and self-correction. Experiments on ACEBench highlight its effectiveness and efficiency, enabling inference-time scaling and reducing output token usage by over 66%. We release data and model checkpoints to facilitate future research.

As multimodal LLM-driven agents continue to advance in autonomy and generalization, evaluation based on static datasets can no longer adequately assess their true capabilities in dynamic environments and diverse tasks. Existing LLM-based synthetic data methods are largely designed for LLM training and evaluation, and thus cannot be directly applied to agent tasks that require tool use and interactive capabilities. While recent studies have explored automatic agent task generation with LLMs, most efforts remain limited to text or image analysis, without systematically modeling multi-step interactions in web environments. To address these challenges, we propose Graph2Eval, a knowledge graph-based framework that automatically generates both multimodal document comprehension tasks and web interaction tasks, enabling comprehensive evaluation of agents' reasoning, collaboration, and interactive capabilities. In our approach, knowledge graphs constructed from multi-source external data serve as the task space, where we translate semantic relations into structured multimodal tasks using subgraph sampling, task templates, and meta-paths. A multi-stage filtering pipeline based on node reachability, LLM scoring, and similarity analysis is applied to guarantee the quality and executability of the generated tasks. Furthermore, Graph2Eval supports end-to-end evaluation of multiple agent types (Single-Agent, Multi-Agent, Web Agent) and measures reasoning, collaboration, and interaction capabilities. We instantiate the framework with Graph2Eval-Bench, a curated dataset of 1,319 tasks spanning document comprehension and web interaction scenarios. Experiments show that Graph2Eval efficiently generates tasks that differentiate agent and model performance, revealing gaps in reasoning, collaboration, and web interaction across different settings and offering a new perspective for agent evaluation.

Recent autonomous penetration testing papers report high benchmark scores while adding multi-component security harnesses around frontier LLMs. Because these systems often change both architecture and backbone model, it is difficult to tell how much performance comes from the harness rather than from the underlying model. This paper presents a controlled study on the 104-task XBOW benchmark using default coding CLI agents as plain-agent baselines. We first run Codex, OpenCode, and Pi with the same GPT-5 model, budget, target interface, and scoring rule. This phase identifies the strongest same-model baseline and tests whether security-specific prompt variants improve its observed score. We then compare the default Codex scaffold with published MAPTA and PentestGPT V2 results under the closest available model matches. Finally, we repeat the plain-agent experiment with GPT-5.2 and GPT-5.5 to measure model scaling inside the same scaffold. The results show a mixed but practical picture. Specialised harnesses can add measurable benchmark lift and may improve cost efficiency, but plain coding agents already solve a large share of the benchmark; repeated plain-agent runs can match or exceed some published architecture scores in union coverage, and newer models substantially improve the same scaffold. Future evaluations should report model-matched plain-agent baselines before attributing benchmark gains to architecture design alone.

Agentic coding harnesses - such as Agent-Skills, Superpowers, and Agent-Rigor - are increasingly deployed to augment underlying LLMs for real-world software engineering tasks. Existing benchmarks evaluate these agents almost exclusively on outcome correctness: whether generated code passes tests or resolves issues. We argue that this outcome-only lens is insufficient: an agent that arrives at a correct solution through reckless trial-and-error, without planning, verification, or graceful recovery, is fundamentally less reliable than one that follows sound engineering discipline. We introduce RigorBench, the first benchmark designed to measure process discipline in AI coding agents. RigorBench evaluates these harnesses across five pillars: Planning Fidelity, Verification Coverage, Recovery Efficiency, Abstention Quality, and Atomic Transition Integrity. A composite RigorScore aggregates these dimensions into a single metric via a weighted sum. We curate a suite of 30 tasks spanning five categories - Plan-Then-Build, Verify-Or-Die, Doom Loop Gauntlet, Know When to Fold, and Don't Break the Build-and evaluate leading harnesses in a controlled with/without experimental design against baseline coding assistants. Our results show that structured process discipline not only improves process quality scores by an average of 41% but also raises downstream outcome correctness by 17%, providing the first quantitative evidence that how agents code matters as much as what they produce. We release the full benchmark, scoring rubrics, and trajectory analysis tools as open-source artifacts.

Recent work pairs LLMs with evolutionary search to iteratively generate, modify, and select code using task-specific feedback. These systems have produced strong results in mathematical discovery and algorithm design, yet a fundamental question remains: what do they actually evolve? Progress is typically summarized by the best score a run reaches under a task-specific evaluator, but that score can reflect several different mechanisms: new algorithmic structure, re-tuning an existing strategy, recombining ideas already in the model's internal knowledge, or overfitting to the evaluator. Distinguishing these mechanisms requires inspecting the search process itself, not only its final outcome. We introduce EvoTrace, a dataset of evolutionary coding traces spanning four evolutionary frameworks, reasoning and non-reasoning models, and 16 tasks across mathematics and algorithm design. To analyze these traces, we develop EvoReplay, a replay-based methodology that reconstructs the local search states behind high-scoring solutions and tests controlled interventions, including adjusting constants, removing program components and substituting models or prompting contexts. We annotate every code edit in EvoTrace with one of nine recurring edit types using an LLM-as-judge pipeline validated against blind human re-annotation. Across EvoTrace, most score gains come from a small subset of these edit types. We further find a deterministic cycling pattern: about 30% of code lines added during search are byte-identical re-introductions of previously-deleted lines, present throughout nearly every run. These results show that benchmark gains in evolutionary coding agents can arise from qualitatively different mechanisms, only some of which correspond to new algorithmic structure. EvoTrace enables more diagnostic evaluation of evolutionary coding agents beyond final benchmark scores.

LLM search agents increasingly rely on tools at inference time, but their trajectories are often constrained by hard limits on both tool calls and generated tokens. Under such dual budgets, better answers require not only stronger models, but also explicit control over which search action should receive the next budget unit and when the accumulated evidence is sufficient to commit a final answer. We study this problem in multi-hop question answering (QA) and formulate it as two-stage inference-time budget control. At search time, our controller assigns each feasible action a task-level Value-of-Information (VOI) score, defined as an operational estimate of marginal task value per unit budget under the current search state and remaining dual budget, and uses this score to choose among retrieval, decomposition, and answer commitment. After search, a selective evidence-grounded finalizer compares the trajectory answer with a refined candidate and rewrites only when the residual error appears to be a low-risk answer-form error. Across four multi-hop QA benchmarks, three LLM backbones, and four budget levels, the method yields positive aggregate gains over four audited baselines under the same hard dual-budget protocol. Ablations show that search-time budget control, especially budget-dependent penalty, provides the main performance gain, while answer-time control helps mainly when the retrieval path is already adequate. These results suggest that inference-time budget control for LLM search agents should govern both how budget is spent during search and how the final answer is committed.

AI agents that interact with users across multiple sessions require persistent long-term memory to maintain coherent, personalized behavior. Current approaches either rely on flat retrieval-augmented generation (RAG), which loses structural relationships between memories, or use memory compression and vector retrieval that cannot capture the associative structure of multi-session conversations. There are few graph based techniques proposed in the literature, however they still suffer from hub dominated retrieval and poor hierarchical reasoning over evolving memory. We propose GAAMA, a graph-augmented associative memory system that constructs a concept-mediated hierarchical knowledge graph through a three-step pipeline: (1)~verbatim episode preservation from raw conversations, (2)~LLM-based extraction of atomic facts and topic-level concept nodes, and (3)~synthesis of higher-order reflections. The resulting graph uses four node types (episode, fact, reflection, concept) connected by five structural edge types, with concept nodes providing cross-cutting traversal paths that complement semantic similarity. Retrieval combines cosine-similarity-based k-nearest neighbor search with edge-type-aware Personalized PageRank (PPR) through an additive scoring function. On the LoCoMo-10 benchmark (1,540 questions across 10 multi-session conversations), GAAMA achieves 78.9\% mean reward, outperforming a tuned RAG baseline (75.0\%), HippoRAG (69.9\%), A-Mem (47.2\%), and Nemori (52.1\%). Ablation analysis shows that augmenting graph-traversal-based ranking (Personalized PageRank) with semantic search consistently improves over pure semantic search on graph nodes (+1.0 percentage point overall).

AI research agents offer the promise to accelerate scientific progress by automating the design, implementation, and training of machine learning models. However, the field is still in its infancy, and the key factors driving the success or failure of agent trajectories are not fully understood. We examine the role that ideation diversity plays in agent performance. First, we analyse agent trajectories on MLE-bench, a well-known benchmark to evaluate AI research agents, across different models and agent scaffolds. Our analysis reveals that different models and agent scaffolds yield varying degrees of ideation diversity, and that higher-performing agents tend to have increased ideation diversity. Further, we run a controlled experiment where we modify the degree of ideation diversity, demonstrating that higher ideation diversity results in stronger performance. Finally, we strengthen our results by examining additional evaluation metrics beyond the standard medal-based scoring of MLE-bench, showing that our findings still hold across other agent performance metrics.

Agentic reinforcement learning (RL) holds great promise for the development of autonomous agents under complex GUI tasks, but its scalability remains severely hampered by the verification of task completion. Existing task verification is treated as a passive, post-hoc process: a verifier (i.e., rule-based scoring script, reward or critic model, and LLM-as-a-Judge) analyzes the agent's entire interaction trajectory to determine if the agent succeeds. Such processing of verbose context that contains irrelevant, noisy history poses challenges to the verification protocols and therefore leads to prohibitive cost and low reliability. To overcome this bottleneck, we propose SmartSnap, a paradigm shift from this passive, post-hoc verification to proactive, in-situ self-verification by the agent itself. We introduce the Self-Verifying Agent, a new type of agent designed with dual missions: to not only complete a task but also to prove its accomplishment with curated snapshot evidences. Guided by our proposed 3C Principles (Completeness, Conciseness, and Creativity), the agent leverages its accessibility to the online environment to perform self-verification on a minimal, decisive set of snapshots. Such evidences are provided as the sole materials for a general LLM-as-a-Judge verifier to determine their validity and relevance. Experiments on mobile tasks across model families and scales demonstrate that our SmartSnap paradigm allows training LLM-driven agents in a scalable manner, bringing performance gains up to 26.08% and 16.66% respectively to 8B and 30B models. The synergizing between solution finding and evidence seeking facilitates the cultivation of efficient, self-verifying agents with competitive performance against DeepSeek V3.1 and Qwen3-235B-A22B.

The rapid growth of scientific literature has made it increasingly difficult for researchers to efficiently discover, evaluate, and synthesize relevant work. Recent advances in multi-agent large language models (LLMs) have demonstrated strong potential for understanding user intent and are being trained to utilize various tools. In this paper, we introduce Paper Circle, a multi-agent research discovery and analysis system designed to reduce the effort required to find, assess, organize, and understand academic literature. The system comprises two complementary pipelines: (1) a Discovery Pipeline that integrates offline and online retrieval from multiple sources, multi-criteria scoring, diversity-aware ranking, and structured outputs; and (2) an Analysis Pipeline that transforms individual papers into structured knowledge graphs with typed nodes such as concepts, methods, experiments, and figures, enabling graph-aware question answering and coverage verification. Both pipelines are implemented within a coder LLM-based multi-agent orchestration framework and produce fully reproducible, synchronized outputs including JSON, CSV, BibTeX, Markdown, and HTML at each agent step. This paper describes the system architecture, agent roles, retrieval and scoring methods, knowledge graph schema, and evaluation interfaces that together form the Paper Circle research workflow. We benchmark Paper Circle on both paper retrieval and paper review generation, reporting hit rate, MRR, and Recall at K. Results show consistent improvements with stronger agent models. We have publicly released the website at https://papercircle.vercel.app/ and the code at https://github.com/MAXNORM8650/papercircle.

While Large Language Models (LLMs) excel at algorithmic code generation, they struggle with front-end development, where correctness is judged on rendered pixels and interaction. We present ReLook, an agentic, vision-grounded reinforcement learning framework that empowers an agent to close a robust generate--diagnose--refine loop by invoking a multimodal LLM (MLLM) as a tool. During training, the agent uses the MLLM-in-the-loop both as a visual critic--scoring code with screenshots--and as a source of actionable, vision-grounded feedback; a strict zero-reward rule for invalid renders anchors renderability and prevents reward hacking. To prevent behavioral collapse, we introduce Forced Optimization, a strict acceptance rule that admits only improving revisions, yielding monotonically better trajectories. At inference, we decouple the critic and run a lightweight, critic-free self-edit cycle, keeping latency comparable to base decoding while retaining most of the gains. Across three widely used benchmarks, ReLook consistently outperforms strong baselines in vision-grounded front-end code generation, highlighting the benefits of agentic perception, visual rewards, and training-inference decoupling.

Despite advancements in enhancing LLM safety against jailbreak attacks, evaluating LLM defenses remains a challenge, with current methods often lacking explainability and generalization to complex scenarios, leading to incomplete assessments (e.g., direct judgment without reasoning, low F1 score of GPT-4 in complex cases, bias in multilingual scenarios). To address this, we present JAILJUDGE, a comprehensive benchmark featuring diverse risk scenarios, including synthetic, adversarial, in-the-wild, and multilingual prompts, along with high-quality human-annotated datasets. The JAILJUDGE dataset includes over 35k+ instruction-tune data with reasoning explainability and JAILJUDGETEST, a 4.5k+ labeled set for risk scenarios, and a 6k+ multilingual set across ten languages. To enhance evaluation with explicit reasoning, we propose the JailJudge MultiAgent framework, which enables explainable, fine-grained scoring (1 to 10). This framework supports the construction of instruction-tuning ground truth and facilitates the development of JAILJUDGE Guard, an end-to-end judge model that provides reasoning and eliminates API costs. Additionally, we introduce JailBoost, an attacker-agnostic attack enhancer, and GuardShield, a moderation defense, both leveraging JAILJUDGE Guard. Our experiments demonstrate the state-of-the-art performance of JailJudge methods (JailJudge MultiAgent, JAILJUDGE Guard) across diverse models (e.g., GPT-4, Llama-Guard) and zero-shot scenarios. JailBoost and GuardShield significantly improve jailbreak attack and defense tasks under zero-shot settings, with JailBoost enhancing performance by 29.24% and GuardShield reducing defense ASR from 40.46% to 0.15%.