Research map / Architectures and efficiency / Attention and transformers
Sign Language Transformers: Joint End-to-End Sign Language Recognition and Translation
Necati Cihan Camgoz, Oscar Koller, Simon Hadfield, Richard Bowden
This paper tried a common change: Add batch / layer normalization
add batch normalization and ReLU to pretrained sign-recognition CNN features · PHOENIX-2014T test · WER: 33.39 -> 27.62 Improved
What 431 accepted papers report for this change
- 313 improved
- 39 no clear effect
- 13 mixed
- 66 hurt
- MemoryLLM: Plug-n-Play Interpretable Feed-Forward Memory for Transformers ICML 2026 · use independent layer norm per FFN · training convergence · convergenceImproved
- Angel or Demon: Investigating the Plasticity Interventions' Impact on Backdoor Threats in Deep Reinforcement Learning ICML 2026 · apply Layer Normalization in TM-Scratch · Acrobot and MountainCar · BTP: reduces BTP by over 30%Hurt
- MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning ICML 2026 · add layer normalization · MEAL Level 1 · soup deliveredImproved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 changed | Where | Measured | Effect |
|---|---|---|---|
| jointly train sign recognition and translation with CTC gloss supervision | Sign Language Transformers · PHOENIX-2014T test | BLEU-4: 9.58 -> 20.17 | Improved |
| replace the RNN-based text-to-text system with a transformer | Gloss2Text · PHOENIX-2014T test | BLEU-4: 19.26 -> 24.54 | Improved |
| add batch normalization and ReLU to pretrained sign-recognition CNN features (ablation) | Sign2Gloss · PHOENIX-2014T test | WER: 33.39 -> 27.62 | Improved |
| use pretrained sign-recognition CNN features instead of EfficientNet-B7 (ablation) | Sign2Gloss · PHOENIX-2014T test | WER: 47.96 -> 33.39 | Improved |
| increase the SLRT depth from 1 to 3 layers (ablation) | Sign2Gloss · PHOENIX-2014T test | WER: 29.90 -> 24.59 | Improved |
| increase the SLRT depth from 3 to 6 layers (ablation) | Sign2Gloss · PHOENIX-2014T test | WER: 24.59 -> 27.78 | Hurt |
| jointly train with equal recognition and translation loss weights instead of recognition-only training (ablation) | Sign2(Gloss+Text) · PHOENIX-2014T test | WER: 24.59 -> 33.75 | Hurt |
| jointly train with equal recognition and translation loss weights instead of translation-only training (ablation) | Sign2(Gloss+Text) · PHOENIX-2014T test | BLEU-4: 20.17 -> 21.22 | Improved |
| increase recognition loss weight from 1.0 to 5.0 while keeping translation weight at 1.0 (ablation) | Sign2(Gloss+Text) · PHOENIX-2014T test | WER: 33.75 -> 24.49 | Improved |
| increase recognition loss weight from 1.0 to 5.0 while keeping translation weight at 1.0 (ablation) | Sign2(Gloss+Text) · PHOENIX-2014T test | BLEU-4: 21.22 -> 21.80 | Improved |
| increase recognition loss weight from 5.0 to 10.0 while keeping translation weight at 1.0 (ablation) | Sign2(Gloss+Text) · PHOENIX-2014T test | BLEU-4: 21.80 -> 21.32 | Improved |
| increase recognition loss weight from 5.0 to 20.0 while keeping translation weight at 1.0 (ablation) | Sign2(Gloss+Text) · PHOENIX-2014T test | WER: 24.49 -> 25.73 | Hurt |
Closest papers by meaning
- Sign2GPT: Leveraging Large Language Models for Gloss-Free Sign Language Translation (ICLR 2024 · similarity 0.92)
- Bridging Sign and Spoken Languages: Pseudo Gloss Generation for Sign Language Translation (NeurIPS 2025 · similarity 0.92)
- Improving Gloss-free Sign Language Translation by Reducing Representation Density (NeurIPS 2024 · similarity 0.91)
- Lost in Translation, Found in Context: Sign Language Translation with Contextual Cues (CVPR 2025 · similarity 0.91)
- Scaling Sign Language Translation (NeurIPS 2024 · similarity 0.91)
- Lost in Translation, Found in Embeddings: Sign Language Translation and Alignment (NeurIPS 2026 · similarity 0.90)
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.
