Research map / Architectures and efficiency / Attention and transformers
Attention-via-Attention Neural Machine Translation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| replace word-level RNNsearch (30K word vocab) with character-level AvA NMT shallow model | RNNsearch · WMT'15 En-Fr newstest2014 | BLEU: 28.5 -> 33.2 | Helped |
| use shallow AvA NMT instead of char2char | character-level NMT · WMT'15 En-Cs newstest2015 | BLEU: 17.6 -> 19.8 | Helped |
| use shallow AvA NMT instead of hybrid model | En-Cs NMT · WMT'15 En-Cs newstest2015 | chrF3: 46.5 -> 48.3 | Helped |
| use deeper AvA NMT (4,2 layers) instead of shallow | AvA NMT · WMT'15 En-Cs newstest2015 | BLEU: 19.8 -> 20.9 | Helped |
| deep AvA NMT versus GNMT on En-Fr | AvA NMT · WMT'15 En-Fr newstest2014 | BLEU: 39.0 -> 40.2 | Helped |
Closest papers by meaning
- Syntax-Directed Attention for Neural Machine Translation (AAAI 2018 · similarity 0.88)
- Attention is All you Need (NeurIPS 2017 · similarity 0.87)
- Translation Heads: Disentangling meaning from language in LLM-based machine translation (ICML 2026 · similarity 0.87)
- Neural Machine Translation with Joint Representation (AAAI 2020 · similarity 0.86)
- Pay Better Attention to Attention: Head Selection in Multilingual and Multi-Domain Sequence Modeling (NeurIPS 2021 · similarity 0.84)
- Stack Attention: Improving the Ability of Transformers to Model Hierarchical Patterns (ICLR 2024 · similarity 0.83)
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.