Video Edit Paper
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Socratic Skill Learning: Interleaving Teaching and Practice for Transferable Agent Skills
TestPrism: Rethinking Test Evaluation Beyond a Single Reference
-
Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories
Paper • 2606.02060 • Published • 58 -
MMG2Skill: Can Agents Distill In-the-Wild Guides into Self-Evolving Skills?
Paper • 2606.01993 • Published • 16 -
NJU-LINK/DR3-Eval
Viewer • Updated • 100 • 8.35k • 2 -
TVIR: Building Deep Research Agents Towards Text--Visual Interleaved Report Generation
Paper • 2606.02320 • Published • 12
Video Edit Paper
-
Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories
Paper • 2606.02060 • Published • 58 -
MMG2Skill: Can Agents Distill In-the-Wild Guides into Self-Evolving Skills?
Paper • 2606.01993 • Published • 16 -
NJU-LINK/DR3-Eval
Viewer • Updated • 100 • 8.35k • 2 -
TVIR: Building Deep Research Agents Towards Text--Visual Interleaved Report Generation
Paper • 2606.02320 • Published • 12
datasets 17
NJU-LINK/SkillTransferBench
RL Environment • Updated • 707 • 343 • 3
NJU-LINK/MT-Video-Bench
Updated • 307 • 6
NJU-LINK/AVE-Compass-v2
Viewer • Updated • 196 • 500 • 3
NJU-LINK/AVSCapBench
Viewer • Updated • 1.23k • 1.72k • 2
NJU-LINK/OmniCap-IF-54K
Viewer • Updated • 53.9k • 1.32k
NJU-LINK/OmniCap-IF
Viewer • Updated • 480 • 395 • 1
NJU-LINK/CoVEBench
Viewer • Updated • 626 • 817 • 2
NJU-LINK/TELBench
Updated • 338 • 2
NJU-LINK/TVIR-Bench
Viewer • Updated • 100 • 106
NJU-LINK/WebCompass
Viewer • Updated • 933 • 1.93k • 6