Multi-Granularity Optimization for Non-Autoregressive Translation
This addresses a key bottleneck in low-latency machine translation for applications requiring fast inference, though it is incremental as it builds on existing non-autoregressive methods.
The paper tackles the performance deterioration in non-autoregressive machine translation caused by the independence assumption and cross-entropy loss, proposing multi-granularity optimization that collects feedback on translation segments. Experiments on WMT benchmarks show it significantly outperforms baselines, achieving best performance on WMT'16 En-Ro and competitive results on WMT'14 En-De.
Despite low latency, non-autoregressive machine translation (NAT) suffers severe performance deterioration due to the naive independence assumption. This assumption is further strengthened by cross-entropy loss, which encourages a strict match between the hypothesis and the reference token by token. To alleviate this issue, we propose multi-granularity optimization for NAT, which collects model behaviors on translation segments of various granularities and integrates feedback for backpropagation. Experiments on four WMT benchmarks show that the proposed method significantly outperforms the baseline models trained with cross-entropy loss, and achieves the best performance on WMT'16 En-Ro and highly competitive results on WMT'14 En-De for fully non-autoregressive translation.