Research map / Large language models / Instruction and fine-tuning
Fine-Tuning by Curriculum Learning for Non-Autoregressive 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 |
|---|---|---|---|
| Use curriculum fine-tuning instead of direct transfer | FCL-NAT · WMT14 En-De | BLEU: 20.23 -> 21.70 | Helped |
| Use curriculum fine-tuning instead of direct transfer | FCL-NAT · WMT14 De-En | BLEU: 23.16 -> 25.32 | Helped |
| Use curriculum fine-tuning instead of direct transfer | FCL-NAT · IWSLT14 De-En | BLEU: 23.05 -> 26.62 | Helped |
| Add token-level substitution during transfer rather than sentence-level substitution (ablation) | FCL-NAT · IWSLT14 De-En validation | BLEU | Helped |
| Use logarithmic rather than ladder-like or linear pacing (ablation) | FCL-NAT · IWSLT14 De-En validation | BLEU | Helped |
| Use ladder-like rather than linear pacing (ablation) | FCL-NAT · IWSLT14 De-En validation | BLEU | No clear effect |
| Add curriculum transfer before purely non-autoregressive training (ablation) | FCL-NAT · IWSLT14 De-En validation | BLEU | Helped |
| Add curriculum fine-tuning (ablation) | FCL-NAT · WMT14 En-De | BLEU: 20.23 -> 21.70 | Helped |
| Add curriculum fine-tuning without explicit repetitive-token regularization | FCL-NAT · IWSLT14 De-En validation | average consecutive repetitive tokens per sentence: 2.30 -> 0.57 | Helped |
Closest papers by meaning
- Non-Autoregressive Neural Machine Translation (ICLR 2018 · similarity 0.89)
- Rephrasing the Reference for Non-autoregressive Machine Translation (AAAI 2023 · similarity 0.89)
- Latent-Variable Non-Autoregressive Neural Machine Translation with Deterministic Inference Using a Delta Posterior (AAAI 2020 · similarity 0.89)
- A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models (ICLR 2024 · similarity 0.88)
- Teaching Large Language Models to Translate with Comparison (AAAI 2024 · similarity 0.87)
- AEQA-NAT : Adaptive End-to-end Quantization Alignment Training Framework for Non-autoregressive Machine Translation (ICML 2025 · similarity 0.86)
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.