AI Paper Atlas

Research map / Video / Video-language understanding

Make Your Training Flexible: Towards Deployment-Efficient Video Models

Chenting Wang, Kunchang Li, Tianxiang Jiang, Xiangyu Zeng, Yi Wang, Limin Wang

ICCV 2025 · Video › Video-language understanding · approach: Long video understanding · 6 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 group-dynamic (L2, 4 groups) token selection instead of random mask (ablation)Flux-Single · K400 w/ TOtop-1 accuracy: 79.0 -> 80.3Improved
use group-dynamic (L2, 4 groups) token selection instead of random mask (ablation)Flux-Single · SSv2 w/ TOtop-1 accuracy: 65.9 -> 67.3Improved
add Flux tuning with token optimization to InternVideo2-S (ablation)InternVideo2-S · K400, 512 tokenstop-1 accuracy: +7.5Improved
add Global-Local positional embedding (GLPE) (ablation)Flux-Single + ViT · K400 w/ TOtop-1 accuracy: 80.3 -> 81.3Improved
add Dual Patch Normalization (ablation)Flux-Single + ViT · K400 w/ TOtop-1 accuracy: 80.3 -> 81.2Improved
add both FluxViT modules (ablation)Flux-Single · SSv2 w/ TOtop-1 accuracy: 67.3 -> 69.3Improved
replace LPE with RoPE (ablation)Flux-Single + ViT · K400 w/ TOtop-1 accuracy: 81.0 -> 80.7Hurt
train with multiple token counts instead of single (ablation)FluxViT · K400 w/ TOtop-1 accuracy: 81.7 -> 82.4Improved
multi-number pretraining and tuning with alignment (ablation)FluxViT-S · K400, 512 test tokenstop-1 accuracy: 65.8 -> 76.6Improved
add Flux augmentation to pretraining and tuning (ablation)FluxViT-S · K400, 512 test tokenstop-1 accuracy: 62.1 -> 65.8Improved
enlarge Tthres to (2048, 6144) (ablation)Flux-Single FluxViT · K400 8x224top-1 accuracy: 80.5 -> 80.1Hurt
remove varied spatial resolution (ablation)Flux-Single FluxViT · K400 with TOtop-1 accuracy: 81.7 -> 81.4No 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