Harness-S: Learning from Sibling Harnesses for Generalizable Agent Training Paper • 2611.04277 • Published 16 days ago • 71
StreamArena: Toward Continuous, Interactive, and Long-Horizon Agentic Streaming Video Understanding Paper • 2608.05703 • Published Aug 6 • 19
VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation Paper • 2607.28590 • Published Jul 30 • 41
MMSkills: Towards Multimodal Skills for General Visual Agents Paper • 2605.13527 • Published May 14 • 100