AI Paper Atlas

Research map / Video / Video-language understanding

Unmasked Teacher: Towards Training-Efficient Video Foundation Models

Kunchang Li, Yali Wang, Yizhuo Li, Yi Wang, Yinan He, Limin Wang, Yu Qiao

ICCV 2023 · Video › Video-language understanding · approach: Video-language pretraining · 193 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
replace pixel reconstruction with unmasked token alignment (ablation)UMT-B (ViT-B/16) · Kinetics-400top-1 accuracy (%): 78.8 -> 84.9Improved
add pixel reconstruction to unmasked token alignment (ablation)UMT-B (ViT-B/16) · Kinetics-400top-1 accuracy (%): 84.9 -> 83.9Hurt
add masked token recovery (ablation)UMT-B (ViT-B/16) · Kinetics-400top-1 accuracy (%): 84.9 -> 84.6Hurt
use semantic masking instead of random masking (ablation)UMT-B (ViT-B/16) · Kinetics-400top-1 accuracy (%): 84.6 -> 84.9Improved
use semantic masking instead of random masking (ablation)UMT-B (ViT-B/16) · Something-Something V2top-1 accuracy (%): 70.2 -> 70.2No clear effect
use sparse sampling instead of dense sampling (ablation)UMT-B (ViT-B/16) · Kinetics-400top-1 accuracy (%): 84.0 -> 84.9Improved
remove temporal downsampling (ablation)UMT-B (ViT-B/16) · Something-Something V2top-1 accuracy (%): 69.5 -> 70.2Improved
use spatiotemporal attention in student pre-training and fine-tuning (ablation)UMT-B (ViT-B/16) · Something-Something V2top-1 accuracy (%): 54.0 -> 70.3Improved
use CLIP spatial teacher instead of spatiotemporal CLIP-ST teacher (ablation)UMT-B (ViT-B/16) · Kinetics-400top-1 accuracy (%): 83.7 -> 85.2Improved
add mask modeling with spatiotemporal attention student (ablation)UMT-B (ViT-B/16) · Something-Something V2top-1 accuracy (%): 69.1 -> 70.3Improved
use two-stage progressive pre-training instead of one stage (ablation)UMT-B (ViT-B/16) · MSRVTT retrievalaverage recall (%): 57.9 -> 67.0Improved
add unmasked teacher alignment in multimodal pre-training (ablation)UMT-B (ViT-B/16) · MSRVTT retrievalaverage recall (%): 66.9 -> 67.0Improved

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