TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces Paper • 2609.33295 • Published 3 days ago • 56
Capability Self-Assessment: Teaching LLMs to Know Their Limits Paper • 2606.00251 • Published May 29 • 10
GeoMotionGPT: Geometry-Aligned Motion Understanding with Large Language Models Paper • 2601.07632 • Published Jan 12 • 2