Research map / Architectures and efficiency / RNNs and normalization
Non-Autoregressive Neural Machine Translation with Enhanced Decoder Input
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 |
|---|---|---|---|
| Add enhanced target-side decoder inputs using phrase-table lookup or embedding mapping | NART · WMT14 En-De | BLEU: 19.17 -> 24.28 | Helped |
| Add enhanced target-side decoder inputs using phrase-table lookup or embedding mapping | NART · WMT16 En-Ro | BLEU: 29.79 -> 34.51 | Helped |
| Use phrase-table lookup rather than word-table lookup for decoder input | ENAT · WMT14 En-De | BLEU: 19.16 -> 20.33 | Helped |
| Add word-level adversary loss L_adv to sentence-level alignment L_align (ablation) | ENAT Embedding Mapping · IWSLT14 De-En | BLEU: 23.74 -> 24.13 | Helped |
| Add sentence-level alignment loss L_align to word-level adversary loss L_adv (ablation) | ENAT Embedding Mapping · IWSLT14 De-En | BLEU: 23.53 -> 24.13 | Helped |
| Use embedding mapping rather than phrase-table lookup | ENAT · WMT14 En-De | BLEU: 20.26 vs. 20.65; rescoring: 23.22 vs. 24.28 | Mixed |
| Use embedding mapping rather than phrase-table lookup | ENAT · IWSLT14 De-En | BLEU: 25.09 -> 24.13 | Hurt |
| Use embedding mapping rather than phrase-table lookup | ENAT · WMT16 En-Ro | BLEU: 29.85 -> 30.08; rescoring: 34.04 -> 34.51 | Helped |
| Use phrase-table lookup rather than embedding mapping | ENAT · WMT14 De-En | BLEU: 23.02 -> 23.23; rescoring: 26.10 -> 26.67 | Helped |
| Increase decoder-input quality from word-table to phrase-table lookup | ENAT · WMT14 En-De | BLEU: 3.54 -> 6.03; 19.16 -> 20.33 | Helped |
Closest papers by meaning
- Non-Autoregressive Machine Translation with Auxiliary Regularization (AAAI 2019 · similarity 0.93)
- Guiding Non-Autoregressive Neural Machine Translation Decoding with Reordering Information (AAAI 2021 · similarity 0.91)
- Minimizing the Bag-of-Ngrams Difference for Non-Autoregressive Neural Machine Translation (AAAI 2020 · similarity 0.91)
- Non-Autoregressive Neural Machine Translation (ICLR 2018 · similarity 0.91)
- Rephrasing the Reference for Non-autoregressive Machine Translation (AAAI 2023 · similarity 0.91)
- AEQA-NAT : Adaptive End-to-end Quantization Alignment Training Framework for Non-autoregressive Machine Translation (ICML 2025 · 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.