AI Paper Atlas

Research map / Video / Video-language understanding

Scaling the Long Video Understanding of Multimodal Large Language Models via Visual Memory Mechanism

Tao Chen, Kun Zhang, Qiong Wu, Xiao Chen, Chao Chang, Xiaoshuai Sun, Yiyi Zhou, Rongrong Ji

CVPR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 8 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 FlexMem visual memoryLLaVA-Video 7B · TimeScopeaccuracy: 65.0 -> 85.9Improved
add FlexMem visual memoryLLaVA-Video 7B · LVBenchaccuracy: 42.6 -> 51.0Improved
add FlexMem visual memoryLLaVA-OneVision 7B · TimeScopeaccuracy: 56.3 -> 80.5Improved
add FlexMem versus AKS retrieval on Video-MMELLaVA-Video 7B · Video-MME (all)accuracy: 65.3 -> 64.7Hurt
add FlexMem versus AdaReTaKe under 24GB limitLLaVA-Video 7B · LVBench (val), 3090accuracy: 46.8 -> 50.2Improved
combine context and local compression (dual-pathway) vs context only (ablation)FlexMem · LongVideoBench (all)accuracy: 62.5 -> 63.6Improved
combine context and local compression (dual-pathway) vs local only (ablation)FlexMem · LongVideoBench (all)accuracy: 62.6 -> 63.6Improved
use both context and local memory during prefill (ablation)FlexMem · LongVideoBench (all)accuracy: 61.9 -> 63.6Improved
recall only relevant memories at decoding instead of all memory bank (ablation)FlexMem · LongVideoBench (all)accuracy: 59.8 -> 63.6Improved
use smaller clips (8 frames) instead of 32 (ablation)FlexMem · LongVideoBench (all)accuracy: 61.7 -> 63.6Improved
replace encoding-based index with fast MemIndex (ablation)FlexMem · MLVU M-avgaccuracy: 72.4 -> 72.3No clear effect
replace encoding-based index with fast MemIndex (last-token, 3 layers) (ablation)FlexMem · LVBench (val)accuracy: 51.0 -> 46.8Hurt

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