RiskChainBench: A Benchmark for Obfuscated Platform Message Restoration and Evidence-Grounded Web Investigation Paper • 2609.16900 • Published 7 days ago • 42
Don't Mask the Environment: Observation Supervision Changes How Agents Explore Under RL Paper • 2609.20715 • Published 5 days ago • 29
ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments Paper • 2609.19134 • Published 6 days ago • 89
Think Before You Link: Rarity, Reasoning, and Retrieval in Multilingual Entity Linking Paper • 2609.10745 • Published 13 days ago • 22
OpenWAM: An Open, Modular Exploration Towards Systematic World-Action Model Pretraining Paper • 2609.07398 • Published 15 days ago • 77