提出零样本自适应策略，让翻译模型实时决策读写动作。
PsFuture: A Pseudo-Future-based Zero-Shot Adaptive Policy for Simultaneous Machine Translation
- 用伪未来信号设计零样本读写策略，无需额外训练。
- 在多基准上达到强基线性能，延迟与质量平衡优秀。
- 适合追求低延迟实时翻译的场景，尤其兼容现有模型。
同时性机器翻译（SiMT）要求在接收流式源端文本时实时生成目标词元。传统方法通常需要复杂的架构和大量参数配置来训练自适应读写策略，带来高昂的计算与内存开销。本文提出 PsFuture，首个面向 SiMT 的零样本自适应读写策略，使翻译模型可自主决定读写行为而无需额外训练。此外，我们设计了一种新型训练策略 Prefix-to-Full（P2F），专门用于将离线翻译模型适配到 SiMT 场景，充分利用离线模型中固有的双向注意力机制。在多个基准上的实验表明，该零样本策略性能媲美强基线，而 P2F 方法可进一步提升性能，实现翻译质量与延迟之间的优异权衡。
原文摘要 · Abstract (English)
Simultaneous Machine Translation (SiMT) requires target tokens to be generated in real-time as streaming source tokens are consumed. Traditional approaches to SiMT typically require sophisticated architectures and extensive parameter configurations for training adaptive read/write policies, which in turn demand considerable computational power and memory. We propose PsFuture, the first zero-shot adaptive read/write policy for SiMT, enabling the translation model to independently determine read/write actions without the necessity for additional training. Furthermore, we introduce a novel training strategy, Prefix-to-Full (P2F), specifically tailored to adjust offline translation models for SiMT applications, exploiting the advantages of the bidirectional attention mechanism inherent in offline models. Experiments across multiple benchmarks demonstrate that our zero-shot policy attains performance on par with strong baselines and the P2F method can further enhance performance, achieving an outstanding trade-off between translation quality and latency.
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。