AI Paper Atlas

Research map / Video / Video-language understanding

ReQuest: Rethinking-based Question-Aware Frame Selection for Long-Form Video QA

Minkuk Kim, Suyong Yun, Young Kim, Jinyoung Moon, Jinwoo Choi, Seong Tae Kim

ECCV 2026 · Video › Video-language understanding · approach: Video question answering · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ReQuest frame selectionLLaVA-Video 7B · Video-MME w/o subtitles overallaccuracy: 62.6 -> 65.6Improved
add ReQuest frame selectionLLaVA-Video 7B · Video-MME w/o subtitles longaccuracy: 52.2 -> 55.8Improved
add ReQuest frame selectionLLaVA-Video 7B · MLVU m-avgaccuracy: 64.7 -> 71.7Improved
add ReQuest frame selectionQwen3-VL-32B · Video-MME w/o subtitles overallaccuracy: 73.9 -> 75.7Improved
add ReQuest frame selectionQwen3-VL-8B · LongVideoBench valaccuracy: 62.7 -> 66.3Improved
use adaptive NMS instead of greedy NMS (ablation)ReQuest on LLaVA-Video · Video-MME w/o subtitles longaccuracy: 53.3 -> 55.8Improved
use adaptive NMS instead of top-k (ablation)ReQuest on LLaVA-Video · Video-MME w/o subtitles overallaccuracy: 64.0 -> 65.6Improved
add uncertainty routing (1365 selection / 1335 uniform) instead of routing all to selection (ablation)ReQuest on LLaVA-Video · Video-MME w/o subtitles overallaccuracy: 65.3 -> 65.6Improved
route everything to uniform sampling (no selection) (ablation)ReQuest on LLaVA-Video · Video-MME w/o subtitles overallaccuracy: 65.6 -> 62.6Hurt
train selector with MLLM response supervision instead of model-based similarity (ablation)ReQuest selector BLIP · Video-MME w/o subtitles overallaccuracy: 63.6 -> 65.6Improved
use frame-level instead of segment-level inference (ablation)ReQuest on LLaVA-Video · Video-MME w/o subtitles overallaccuracy: 62.9 -> 65.6Improved
replace uniform re-inference with selection re-inference (ablation)Router-Ours · Video-MME w/o subtitles overallaccuracy: 62.8 -> 65.6Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy