English

Non-autoregressive Streaming Transformer for Simultaneous Translation

Computation and Language 2023-10-24 v1 Artificial Intelligence

Abstract

Simultaneous machine translation (SiMT) models are trained to strike a balance between latency and translation quality. However, training these models to achieve high quality while maintaining low latency often leads to a tendency for aggressive anticipation. We argue that such issue stems from the autoregressive architecture upon which most existing SiMT models are built. To address those issues, we propose non-autoregressive streaming Transformer (NAST) which comprises a unidirectional encoder and a non-autoregressive decoder with intra-chunk parallelism. We enable NAST to generate the blank token or repetitive tokens to adjust its READ/WRITE strategy flexibly, and train it to maximize the non-monotonic latent alignment with an alignment-based latency loss. Experiments on various SiMT benchmarks demonstrate that NAST outperforms previous strong autoregressive SiMT baselines.

Keywords

Cite

@article{arxiv.2310.14883,
  title  = {Non-autoregressive Streaming Transformer for Simultaneous Translation},
  author = {Zhengrui Ma and Shaolei Zhang and Shoutao Guo and Chenze Shao and Min Zhang and Yang Feng},
  journal= {arXiv preprint arXiv:2310.14883},
  year   = {2023}
}

Comments

EMNLP 2023 main conference; Source code is available at https://github.com/ictnlp/NAST

R2 v1 2026-06-28T12:58:53.602Z