SoftwareGym: Scaling Computer-Use Agent Training with Diverse Synthesized Software Environments Paper • 2610.12611 • Published 7 days ago • 101
view article Article autotrust/JEV-27B: fast, calibrated decisions and full reasoning from one open model autotrust • 18 days ago • 48
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published 15 days ago • 664