DeLoop: Repeated Actions as Self-Supervision for Long-Horizon Agents Paper • 2610.19376 • Published 23 days ago • 21
DTDD: Divergence-Triggered Dynamic Distillation for Reliable On-Policy Supervision Paper • 2610.19302 • Published 24 days ago • 5
CLASHarness: Benchmarking Multi-Agent Conflict in Computer Use via a Controllable and Verifiable Generation Pipeline Paper • 2610.19128 • Published 24 days ago • 34