Pack Light, CarryOn: Building Object-Oriented Context for Long-Horizon Agents Paper • 2610.18784 • Published Oct 13 • 118
TimeEvo: Failure-Driven Self-Evolution of a Time Series Agent Paper • 2609.27277 • Published Sep 23 • 28
TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces Paper • 2609.33295 • Published Sep 27 • 62
RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents Paper • 2609.22000 • Published Sep 18 • 73
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself Paper • 2609.22068 • Published Sep 18 • 117
ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement Paper • 2609.14857 • Published Sep 14 • 55
Confidence Comes from Experience: Experiential Confidence Estimation from Reasoning to Agents Paper • 2609.17708 • Published Sep 15 • 46
RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments Paper • 2609.15364 • Published Sep 14 • 89
SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking Paper • 2609.13141 • Published Sep 11 • 66
AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnesses Paper • 2608.12307 • Published Aug 12 • 107
OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models Paper • 2607.28609 • Published Jul 30 • 69
LLMRouter: Unified Infrastructure for Developing, Evaluating, and Deploying LLM Routers Paper • 2608.06867 • Published Aug 7 • 90
Graph of Skills: Dependency-Aware Structural Retrieval for Massive Agent Skills Paper • 2604.05333 • Published Apr 7 • 21
EvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification Paper • 2604.01687 • Published Apr 2 • 8