Research map / Architectures and efficiency / RNNs and normalization
Syntactic Skeleton-Based 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 |
|---|---|---|---|
| add syntactic skeleton with partially syntactic rules to hierarchical phrase-based | Hierarchical Phrase-based · zh-en nw, zh-en web, en-zh tests | BLEU: +0.95 average (binary trees) | Helped |
| use binary parse trees instead of CTB-style trees for SYNSKEL (ablation) | SYNSKEL · zh-en nw, zh-en web, en-zh tests | BLEU: +0.71 -> +0.95 | Helped |
| use tree-parsing instead of string-parsing (ablation) | SYNSKEL · zh-en nw, zh-en web, en-zh tests | BLEU: +0.95 -> +0.91 | No clear effect |
| use binary forest instead of single binary tree (ablation) | SYNSKEL · zh-en nw, zh-en web, en-zh tests | BLEU: +0.91 -> +0.94 | No clear effect |
| SYNSKEL instead of Hier. + SAMT rules | SYNSKEL · zh-en nw, zh-en web, en-zh tests | BLEU: +0.35 -> +0.95 | Helped |
| limit maximum skeleton depth to 5 or less instead of unlimited (ablation) | SYNSKEL · zh-en newswire tuning | BLEU and decoding time | Mixed |
| use tree-to-string model alone instead of hierarchical phrase-based | Tree-to-String · zh-en nw, zh-en web, en-zh tests | BLEU: -0.77 (CTB), -0.39 (binary) | Hurt |
| add source syntax features to hierarchical phrase-based | Hierarchical Phrase-based · zh-en nw, zh-en web, en-zh tests | BLEU: +0.23 (CTB), +0.29 (binary) | Helped |
Closest papers by meaning
- Unsupervised Machine Translation Using Monolingual Corpora Only (ICLR 2018 · similarity 0.83)
- Controlling Neural Machine Translation Formality with Synthetic Supervision (AAAI 2020 · similarity 0.83)
- Exploring the Translation Mechanism of Large Language Models (NeurIPS 2025 · similarity 0.82)
- A Benchmark for Learning to Translate a New Language from One Grammar Book (ICLR 2024 · similarity 0.82)
- A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models (ICLR 2024 · similarity 0.81)
- MT-Ranker: Reference-free machine translation evaluation by inter-system ranking (ICLR 2024 · similarity 0.81)
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.