AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Visual Alignment Pre-training for Sign Language Translation

Peiqi Jiao, Yuecong Min, Xilin Chen

ECCV 2024 · Large language models › Language, safety and interpretability · approach: Sign and cross-lingual translation · 22 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
add VAP visual alignment pre-training to skeleton-based SLT baselineCoSign-1s baseline · PHOENIX14T TestBLEU-4: 19.72 -> 26.16Improved
add VAP to baselineCoSign-1s baseline · OpenASL TestBLEU-4: 10.52 -> 21.23Improved
re-initialize the translation module of baseline once (ablation)gloss-free SLT baseline · PHOENIX14TBLEU-4: 19.72 -> 23.04 (Test)Improved
re-initialize translation module more times (iter 3) (ablation)gloss-free SLT baseline · PHOENIX14T DevBLEU-4: 23.75 -> 22.61Mixed
pre-train visual encoder with L_Align (ablation)gloss-free SLT baseline · PHOENIX14T TestBLEU-4: 19.72 -> 22.94Improved
add L_SLT to L_Align pre-training (ablation)VAP pre-training · PHOENIX14T TestBLEU-4: 22.94 -> 24.62Improved
fine-tune with mBART pre-trained translation module instead of from scratch (ablation)VAP · PHOENIX14T DevBLEU-4: 24.38 -> 26.23Improved
add pseudo-gloss-to-text (P2T) fine-tuning of mBART (ablation)VAP · PHOENIX14T DevBLEU-4: 26.23 -> 26.48No clear effect
keep L_Align during end-to-end fine-tuning (ablation)VAP · PHOENIX14T DevBLEU-4: 26.48 -> 26.62No clear effect
use max pooling instead of average pooling for similarity aggregation (ablation)VAP · PHOENIX14T TestBLEU-4: 17.13 -> 25.54Improved
replace pre-trained embedding layer with Transformer text encoder (ablation)VAP text encoder · PHOENIX14T DevBLEU-4: 26.62 -> 25.52Hurt
use pseudo glosses with embedding-initialized CTC in place of real glosses (ablation)SLT trained from scratch · PHOENIX14T TestBLEU-4: 26.14 vs 27.00 with glossMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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