给大模型翻译加新动作，让机器口译更像人。
Redefining Machine Simultaneous Interpretation: From Incremental Translation to Human-Like Strategies
- 引入四种新操作：切句、删减、简写和代词化，动态调整输出
- 在英中/英德双语任务上，延迟更低且语义质量更高
- 适合追求实时性与自然度的语音翻译系统开发者
同步机器翻译（SiMT）需在严格实时约束下生成高质量译文，传统编码器-解码器框架仅用读取/书写动作难以应对。本文扩展动作空间，引入四种自适应动作：句子切分（SENTENCE_CUT）、删减（DROP）、部分摘要（PARTIAL_SUMMARIZATION）和代词化（PRONOMINALIZATION），实现实时重构、省略与简化，同时保持语义一致。在仅解码器架构的大语言模型中实现这些动作，并通过动作感知提示构建训练参考。为评估质量和延迟，设计了考虑延迟的语音合成流水线，使文本输出映射为具真实时间特征的语音。在ACL60/60英中和英德基准测试中，本框架持续提升语义指标（如COMET-KIWI），并实现更低平均滞后延迟，优于参考译文和萨拉米基基线。值得注意的是，结合DROP与SENTENCE_CUT可达到流畅性与延迟的最佳平衡。结果表明，丰富基于大模型的同步翻译动作空间，是缩小机器与人类口译差距的可行方向。
原文摘要 · Abstract (English)
Simultaneous Machine Translation (SiMT) requires high-quality translations under strict real-time constraints, which traditional encoder-decoder policies with only READ/WRITE actions cannot fully address. We extend the action space of SiMT with four adaptive actions: SENTENCE_CUT, DROP, PARTIAL_SUMMARIZATION and PRONOMINALIZATION, which enable real-time restructuring, omission, and simplification while preserving semantic fidelity. We implement these actions in a decoder-only large language model (LLM) framework and construct training references through action-aware prompting. To evaluate both quality and latency, we further develop a latency-aware TTS pipeline that maps textual outputs to speech with realistic timing. Experiments on the ACL60/60 English-Chinese and English-German benchmarks show that our framework consistently improves semantic metrics (e.g., COMET-KIWI) and achieves lower delay (measured by Average Lagging) compared to reference translations and salami-based baselines. Notably, combining DROP and SENTENCE_CUT yields the best overall balance between fluency and latency. These results demonstrate that enriching the action space of LLM-based SiMT provides a promising direction for bridging the gap between human and machine interpretation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。