{"ID":5675146,"CreatedAt":"2026-07-03T01:40:09.565152011Z","UpdatedAt":"2026-07-05T05:45:35.603470622Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.01737","arxiv_id":"2607.01737","title":"ReQuest: Rethinking-based Question-Aware Frame Selection for Long-Form Video QA","abstract":"Recent multimodal large language models (MLLMs) have substantially advanced video understanding, yet long-form video QA remains challenging under fixed input token budgets, where uniform sampling can be inefficient for evidence localization. We propose ReQuest , an uncertainty-driven, question-adaptive keyframe selection pipeline that aligns question intent with relevant video content through selective computation. ReQuest integrates (i) a lightweight question-aware selector distilled from MLLM-generated supervision, (ii) Re-thinking Routing that triggers additional inference only when the model is uncertain with a length-adaptive criterion, and (iii) uncertainty-guided adaptive non-maximum suppression that selects temporally diverse frames while adjusting spacing based on question difficulty. As a plug-andplay method, ReQuest improves long-video QA without modifying or fine-tuning the underlying MLLM. Experiments on Video-MME, MLVU, and LongVideoBench demonstrate consistent accuracy gains with competitive computational cost, with particularly strong improvements in medium and long video regimes.","short_abstract":"Recent multimodal large language models (MLLMs) have substantially advanced video understanding, yet long-form video QA remains challenging under fixed input token budgets, where uniform sampling can be inefficient for evidence localization. We propose ReQuest , an uncertainty-driven, question-adaptive keyframe selecti...","url_abs":"https://arxiv.org/abs/2607.01737","url_pdf":"https://arxiv.org/pdf/2607.01737v1","authors":"[\"Minkuk Kim\",\"Suyong Yun\",\"Young Tae Kim\",\"Jinyoung Moon\",\"Jinwoo Choi\",\"Seong Tae Kim\"]","published":"2026-07-02T05:46:42Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}