AI Paper Atlas

Research map / Video / Video-language understanding

ReMoRa: Multimodal Large Language Model based on Refined Motion Representation for Long-Video Understanding

Daichi Yashima, Shuhei Kurita, Yusuke Oda, Komei Sugiura

CVPR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 5 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
use codec-aware 64 I-frames instead of uniform 64 frames (ablation)ReMoRa · VideoMMEaccuracy: 62.4 -> 64.3Improved
increase number of I-frames from 16 to 64 (ablation)ReMoRa · VideoMMEaccuracy: 58.5 -> 64.3Improved
increase number of I-frames from 16 to 64 (ablation)ReMoRa · NExT-QAaccuracy: 81.6 -> 84.2Improved
add optical-flow pretraining of RMR (ablation)ReMoRa · NExT-QAaccuracy: 82.2 -> 84.2Improved
add RMR module (ablation)ReMoRa · VideoMMEaccuracy: 62.1 -> 64.3Improved
use HMSS instead of cross-attention fusion (ablation)ReMoRa · VideoMMEaccuracy: 62.5 -> 64.3Improved
use HMSS instead of additive fusion (ablation)ReMoRa · NExT-QAaccuracy: 81.5 -> 84.2Improved
use ReMoRa instead of LLaVA-Video for memory usevideo MLLM inference · NExT-QA 50 videos (H200)peak GPU memory (GB): 23.21 -> 10.59Improved
use ReMoRa instead of LLaVA-Video for throughputvideo MLLM inference · NExT-QA 50 videos (H200)samples/s: 0.53 -> 0.40Hurt
re-encode with harsh H.264 settings (CRF 35, GOP 64, no B-frames, 256p) (ablation)ReMoRa · NExT-QAaccuracy: 84.9 -> 81.5Hurt
vary GOP size or B-frames in H.264 encoding (ablation)ReMoRa · NExT-QAaccuracy: 84.9 -> 84.3 to 84.6No clear effect

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