Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories Paper • 2606.02060 • Published 20 days ago • 55 • 9
CoVEBench: Can Video Editing Models Handle Complex Instructions? Paper • 2606.08415 • Published 14 days ago • 49
OmniCap-IF: Benchmarking and Improving Instruction Following Abilities for Omni-Video Captioning Paper • 2606.08572 • Published 14 days ago • 14
CoVEBench: Can Video Editing Models Handle Complex Instructions? Paper • 2606.08415 • Published 14 days ago • 49 • 3
CoVEBench: Can Video Editing Models Handle Complex Instructions? Paper • 2606.08415 • Published 14 days ago • 49
CoVEBench: Can Video Editing Models Handle Complex Instructions? Paper • 2606.08415 • Published 14 days ago • 49
Socratic-SWE: Self-Evolving Coding Agents via Trace-Derived Agent Skills Paper • 2606.07412 • Published 16 days ago • 12
Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories Paper • 2606.02060 • Published 20 days ago • 55 • 9