OSWorld-Pro: Process-based Evaluation for Computer Use Agents Paper • 2609.24890 • Published 23 days ago • 28
Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents Paper • 2609.39982 • Published 15 days ago • 119
Agora: Git as Shared Memory for Collective AutoResearch Paper • 2609.18094 • Published 29 days ago • 59