AI RESEARCH
Where to Focus: Query-Modulated Multimodal Keyframe Selection for Long Video Understanding
arXiv CS.CV
•
ArXi:2604.17422v1 Announce Type: new Long video understanding remains a formidable challenge for Multimodal Large Language Models (MLLMs) due to the prohibitive computational cost of processing dense frame sequences. Prevailing solutions, which select a keyframe subset, typically rely on either a single visual-centric metric (e.g., CLIP similarity) or a static fusion of heuristic scores. This ``one-size-fits-all'' paradigm frequently fails: visual-only metrics are ineffective for plot-driven narrative queries, while indiscriminately incorporating textual scores.