AI Paper Atlas

Research map / Video / Video-language understanding

Seeing the Scene Matters: Revealing Forgetting in Video Understanding Models with a Scene-Aware Long-Video Benchmark

Seng Nam Chen, Hao Chen, Chenglam Ho, Xinyu Mao, Jinping Wang, Yu Zhang, Chao Li

CVPR 2026 · Video › Video-language understanding · approach: Long video understanding · 0 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
move from clip-level to scene-level questionsopen-source MLLMs · SceneBenchaccuracy: ClipQA 51.73 -> SceneQA 26.11Hurt
add Scene-RAG to base MLLMsLongVA, Long-LLaVA, LLaVA-OneVision · SceneBenchavg accuracy: 53.3 -> 54.4 (LongVA); 52.9 -> 54.6 (Long-LLaVA); 58.1 -> 59.4 (LLaVA-OV)Improved
add frame-level Video-RAG to base MLLMsLongVA, Long-LLaVA, LLaVA-OneVision · SceneBenchavg accuracy: +0.1 / +0.0 / +0.2No clear effect
add Video-RAG to LLaVA-OneVision on I-VQALLaVA-OneVision-7B · SceneBench I-VQAaccuracy: 46.5 -> 43.7Hurt
add Scene-RAG to base MLLMsLongVA, Long-LLaVA, LLaVA-OneVision · Video-MME overallaccuracy: 51.4 -> 63.4 (LongVA); 52.0 -> 64.8 (Long-LLaVA); 43.0 -> 50.6 (LLaVA-OV)Improved
add visual retrieval only (ablation)LongVA · SceneBenchavg accuracy: 53.3 -> 53.5No clear effect
add audio retrieval (ablation)LongVA Scene-RAG · SceneBenchavg accuracy: 53.5 -> 54.0Improved
add scene tiling (ablation)LongVA Scene-RAG · SceneBenchavg accuracy: 54.0 -> 54.4Improved
add audio retrieval (ablation)LLaVA-Video-7B Scene-RAG · MLVUaccuracy: 71.2 -> 73.4Improved
add scene tiling (ablation)LLaVA-Video-7B Scene-RAG · MLVUaccuracy: 73.4 -> 74.1Improved
add audio track as input (SceneQA-Audio vs SceneQA)open-source MLLMs · SceneBenchaccuracy: 26.11 -> 29.08Improved

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