01. GitHarness: Git Init Your Harness Working Memory for Perpetual User Requirements

84% worth reading · novelty 2.0/3 · new method · cs.MA

02. Do LLM Agents Execute the Plans They Declare? From Planning-Mode Declaration to Pattern-Specific Execution

86% worth reading · novelty 1.6/3 · new method · cs.AI

03. Thinking Before Thinking: Scaling Agentic Inference Through Meta-Reasoning

84% worth reading · novelty 1.7/3 · new method · cs.AI

04. Rational Clarification by Assistive Agents via Value-of-Information Reasoning

82% worth reading · novelty 1.7/3 · new method · cs.MA

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

79% worth reading · novelty 1.9/3 · new method · cs.AI

06. AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation

81% worth reading · novelty 1.7/3 · new method · cs.AI

07. An LLM-powered Agent Framework for Heterogeneous Evacuation Behavior Modeling under a Moving Threat in a Public Plaza

75% worth reading · novelty 2.0/3 · new method · cs.MA

08. FlowMAS: Learning Multi-Agent Workflow Topology via Information-guided Generative Flow Network

78% worth reading · novelty 1.7/3 · new method · cs.MA

09. Skill-Space Shooting for Autonomous Robot Policy Improvement

73% worth reading · novelty 1.9/3 · new method · cs.AI

10. Correct Answers, Invalid Traces: What Verifiable Grade-School Math Reveals About Chain-of-Thought Traces

75% worth reading · novelty 1.7/3 · benchmark/eval · cs.AI

11. LLM-Based Multi-Agent Systems over Wireless Networks: A Joint Agent--Network Design Perspective

66% worth reading · novelty 2.0/3 · new method · cs.MA

12. Character Training for Risk-Averse Agents

66% worth reading · novelty 1.7/3 · new method · cs.AI

13. LongHarness Bench: Stress-Testing Language Model Harnesses for Long-Context Reasoning

64% worth reading · novelty 1.7/3 · benchmark/eval · cs.CL

14. Beyond the Timeline: Augmenting Long-Video Memory with Grounded Entity Biographies

60% worth reading · novelty 1.7/3 · new method · cs.AI

15. Pretraining Latent Information Feedback Transformers with Teacher Supervision

54% worth reading · novelty 2.0/3 · new method · cs.CL

16. From Routing Signals to Selective Review: Visual regrounding in MoE VLMs

52% worth reading · novelty 1.9/3 · new method · cs.CL

17. Cooperative Multi-Agent Vision-Language-Action Models via Reinforced Fine Tuning

56% worth reading · novelty 1.7/3 · new method · cs.MA

18. Human-AI Collaboration: From Paradoxes to Patterns

59% worth reading · novelty 1.1/3 · survey/position · cs.MA

19. A Spectral Theory of Distortion in LLM Graph Reconstruction: Sharp Bounds and Empirical Characterization

44% worth reading · novelty 2.0/3 · benchmark/eval · cs.LG

20. IMPACT: Modeling Socially Interdependent Movement in a Generative Multi-Agent Simulation of a Pompeian Household

43% worth reading · novelty 1.9/3 · new method · cs.MA