AI Paper Atlas

Research map / Video / Video-language understanding

Video-Text Alignment Model for Sign Language Translation

Junyi Hu, Zhewen He, Haomian Huang, Yi Fang, Aoxiang Yang

ECCV 2026 · Video › Video-language understanding · approach: Video-language 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
add position-aligned contrastive loss (ablation)VTaMo · Phoenix-2014T testBLEU-4: 22.56 -> 28.86Improved
add local OT alignment loss (ablation)VTaMo · Phoenix-2014T testBLEU-4: 23.77 -> 28.86Improved
add global EMD loss (ablation)VTaMo · Phoenix-2014T testBLEU-4: 26.21 -> 28.86Improved
add orthogonality constraint on global transform (ablation)VTaMo · Phoenix-2014T testBLEU-4: 28.13 -> 28.86Improved
add memory queue to global alignment (ablation)VTaMo · Phoenix-2014T testBLEU-4: 27.35 -> 28.86Improved
add scheduled activation of global loss (ablation)VTaMo · Phoenix-2014T testBLEU-4: 26.42 -> 28.86Improved
use attention pooling instead of max-pooling in temporal encoder (ablation)VTaMo · Phoenix-2014T testBLEU-4: 26.57 -> 28.86Improved
use attention pooling instead of max-pooling in temporal encoder (ablation)VTaMo · Phoenix-2014T testROUGE-L: 55.23 -> 60.24Improved
replace VTaMo alignment losses with generic CLIP4Clip/VideoCLIP/X-CLIP objectives (ablation)VTaMo pipeline · Phoenix-2014T testBLEU-4: 28.86 -> 9.47Hurt
add VTaMo alignment losses to CLIP-ViT-only pipeline (ablation)VTaMo pipeline · Phoenix-2014T testBLEU-4: 10.45 -> 28.86Improved
add learnable null token to OT alignment (ablation)VTaMo local alignment · Phoenix-2014T test (dev)alignment peak probability: 0.61 -> 0.76Improved
add epsilon annealing to Sinkhorn (ablation)VTaMo local alignment · Phoenix-2014T test (dev)alignment entropy: 0.55 -> 0.18Improved

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