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.

提出了一种非自回归流式 Transformer（NAST）用于同时机器翻译（SiMT），通过通过一种新的编码器和非自回归解码器拓宽了读 / 写策略并降低延迟损失，实验证明 NAST 优于以往自回归 SiMT 基线模型。