Research map / Large language models / Language, safety and interpretability
SASST: Leveraging Syntax-Aware Chunking and LLMs for Simultaneous Speech Translation
Zeyu Yang, Lai Wei, Roman Koshkin, Xi Chen, Satoshi Nakamura
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| use syntax-aware chunking instead of fixed-length segmentation (ablation) | SASST · CoVoST2 En-Zh | BLEU: 23.2 -> 38.5 | Improved |
| use syntax-aware chunking instead of fixed-length segmentation (ablation) | SASST · CoVoST2 En-Zh | boundary alignment: 23% -> 82% | Improved |
| use Qwen3-8B instead of LLaMA3-8B backbone (ablation) | SASST · CoVoST2 En-Zh | BLEU: 37.048 -> 40.216 | Improved |
| use Qwen3-8B instead of LLaMA3-8B backbone (ablation) | SASST · CoVoST2 En-De | BLEU: 26.684 -> 27.892 | Improved |
| use Qwen3-8B instead of LLaMA3-8B backbone (ablation) | SASST · CoVoST2 En-Ja | BLEU: 25.674 -> 27.279 | Improved |
| increase input chunk size (latency from ~1.8 s to ~4.0 s) (ablation) | SASST · CoVoST2 En-Zh | BLEU: 34.1 -> 41.5 | Improved |
Closest papers by meaning
- High-Fidelity Simultaneous Speech-To-Speech Translation (ICML 2025 · similarity 0.87)
- Divergence-Guided Simultaneous Speech Translation (AAAI 2024 · similarity 0.86)
- A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models (ICLR 2024 · similarity 0.85)
- Training-Free Long-Context Scaling of Large Language Models (ICML 2024 · similarity 0.84)
- Chunk-Distilled Language Modeling (ICLR 2025 · similarity 0.84)
- Efficient Long Context Fine-tuning with Chunk Flow (ICML 2025 · similarity 0.84)
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.
