new

Get trending papers in your email inbox once a day!

Get trending papers in your email inbox!

Daily Papers

byAK and the research community

Oct 1

Submitted by

ziyjiang

WorldAuditBench: Interactive 3D World Auditing with Multimodal Agents

UCSantaBarbara

University of California, Santa Barbara

Submitted by

chengq9

Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI

apodex

Submitted by

qizekun

Systematically Exploring the Capabilities of GPT-6 Astra as Embodied Policies

Galbot

Submitted by

LH2101

The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation

·
7 authors

Submitted by

passing2961

EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery

minnesotanlp

Submitted by

JEdward

RoboCoach: World Models as Active Coaches for Compositional Robot Skills

Tsinghua University

Submitted by

zhiyuan218

More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Models

JilinUniversity001

Jilin University

Submitted by

henry-y1

DC-SAE: Deep Compression Semantic Autoencoder for Faster Diffusion Convergence

DAGroup-PKU

Submitted by

donghao-zhou

ThinkV2V: Unleashing the Reasoning Capability of MLLMs for Instruction-Guided Video Editing

·
13 authors

Submitted by

imhgchoi

AIM: Agentic Idea Management for Automated Research

google

Submitted by

WenyiWU0111

RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement

RSIGame

Submitted by

henggg

DuoOPD: Learning from Joint Teacher-Student Outcomes for Multi-Task On-Policy Distillation

·
9 authors

Submitted by

Yesianrohn

Decompose Radicals, Then Reward: Fine-Grained Inspection for Accurate Chinese Text Rendering

Fudan-University

Fudan University

Submitted by

Glanceyes

A2Z GameSpec-Bench: How Faithfully Can Coding Agents Generate Games from Game Design Specifications?

KRAFTON

Submitted by

ElvisWang111

Thinking Outside the Box: Can Language Models Rely on External Guidance Selectively?

·
5 authors

Submitted by

taesiri

Game-Guided Skill Discovery through Self-Play for Playable Agent Control

·
4 authors

Submitted by

taesiri

Physis-Lang: Self-Evolving Language as a Physical Representation for Video World Model

nvidia

Submitted by

Leozkl

Agent Error Dataset: Scaling 50,000 Error--Diagnosis Pairs for Failure Analysis and Error-Aware Post-Training

illinois

University of Illinois at Urbana-Champaign

Submitted by

HeiXiong620

CoEvoWhen: Policy-Tool Coevolution for Ultra-Long Video Temporal Grounding

zju

Zhejiang University

Submitted by

bunny127

EviRover: Reinforcing Agentic Perception Beyond a Glance

·
7 authors

Submitted by

Luli3220

Not Every Token Is Worth Distilling: Selective Supervision for Direct-OPD

ECNU

East China Normal University

Submitted by

softmax

WUSH-KV: KV Cache Quantization with Data-Adaptive Transforms

ISTA-DASLab

IST Austria Distributed Algorithms and Systems Lab

1

Submitted by

rzdiversity

See it, Say it, Sorted: Mechanistic Diagnosis and Parameter-Space Mitigation of Emergent Misalignment in LLMs

UniversityofBirmingham

University of Birmingham

Submitted by

HwanChang0106

Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering

kaist-ai

Submitted by

guanqun-yang

SkillSeek: Revisiting Agent Skill Retrieval at Marketplace Scale

StevensAGI

Stevens Institute of Technology

Submitted by

guanqun-yang

PatchHolmes: Agentic Patch Retrieval via Listwise Selection

StevensAGI

Stevens Institute of Technology

Submitted by

lorenzomazza

Understanding Multimodality in Generative Behavioral Cloning

nct-tso

Department of Translational Surgical Oncology (TSO) at NCT Dresden

Submitted by

bhavikmangla

Almost Human, Except When It Matters: VoxParity and the Decisions a Voice Should Change

·
1 authors