StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling Paper • 2608.15089 • Published 22 days ago • 446
VideoCoCo: Code-as-CoT for Physically-Consistent Video Generation via an Agentic Dual-Engine System Paper • 2607.27380 • Published Jul 29 • 72
JarvisHub: An Open Harness for Canvas-Native Multimodal Creative Agents Paper • 2607.23588 • Published Jul 26 • 126
Rewarding the Rare: Uniqueness-Aware RL for Creative Problem Solving in LLMs Paper • 2601.08763 • Published Jan 13 • 150