StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling Paper • 2608.15089 • Published 8 days ago • 436
ClawGUI: A Unified Framework for Training, Evaluating, and Deploying GUI Agents Paper • 2604.11784 • Published Apr 13 • 143
All Roads Lead to Rome: Incentivizing Divergent Thinking in Vision-Language Models Paper • 2604.00479 • Published Apr 1 • 70