AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Scaling up Multimodal Pre-Training for Sign Language Understanding

Wengang Zhou, Weichao Zhao, Hezhen Hu, Zecheng Li, Houqiang Li

TPAMI 2025 · Vision-language and multimodal › CLIP and zero-shot · approach: Recaptioning and text-supervised pretraining

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
increase SL-1.5M pre-training data from 0% to 100% (ablation)Ours · MSASL1000Top-1 accuracy: 62.23 -> 74.07Improved
increase SL-1.5M pre-training data from 0% to 100% (ablation)Ours · Phoenix14WER: 28.9 -> 21.2Improved
increase SL-1.5M pre-training data from 0% to 100% (ablation)Ours · Phoenix14-TBLEU-4: 13.15 -> 22.24Improved
add sign-text contrastive learning to masked pose modeling (ablation)Ours · MSASL1000Top-1 accuracy: 69.44 -> 74.07Improved
add masked pose modeling to sign-text contrastive learning (ablation)Ours · MSASL1000Top-1 accuracy: 67.16 -> 74.07Improved
add non-manual (face and mouth) features to manual features (ablation)Ours · MSASL1000Top-1 accuracy: 70.24 -> 74.07Improved
use fine-grained sign-text similarity instead of coarse-grained (ablation)Ours · CSL-Daily SL-RTT2V R@1: 79.8 -> 87.5Improved
use GCN pose embedding instead of linear projector (ablation)Ours · CSL-Daily SL-RTT2V R@1: 77.4 -> 87.5Improved
set mask ratio to 40% instead of 0% (ablation)Ours · MSASL1000Top-1 accuracy: 67.16 -> 74.07Improved
raise mask ratio from 40% to 80% (ablation)Ours · MSASL1000Top-1 accuracy: 74.07 -> 66.36Hurt
enlarge sliding window from 4 to 16 (ablation)Ours · Phoenix14-TBLEU-4: 22.24 -> 18.17Hurt
set contrastive loss weight to 1.0 instead of 0.1 (ablation)Ours · MSASL1000Top-1 accuracy: 71.74 -> 74.07Improved

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