WorldAuditBench: Interactive 3D World Auditing with Multimodal Agents Paper • 2609.40325 • Published 3 days ago • 95
TIPS: Turn-Level Information-Potential Reward Shaping for Search-Augmented LLMs Paper • 2603.22293 • Published Mar 11 • 1
IsoCompute Playbook: Optimally Scaling Sampling Compute for LLM RL Paper • 2603.12151 • Published Mar 12 • 2