arXivMichal Shlapentokh-Rothman, Prachi Garg, Yu-Xiong Wang, Derek HoiemFri, May 22, 2026, 9:29 AM PDT
score 14.7
AI breaks down video questions to find relevant scenes faster
Original: Decomposing Queries into Tool Calls for Long-Video Keyframe Retrieval
Source: arxiv.org ↗
Writing ELI5 summary…