Uncertainty-Aware Semantic Augmentation for Neural Machine Translation
This addresses a key problem in neural machine translation for improving translation quality by reducing distribution mismatch, though it appears incremental as it builds on existing augmentation techniques.
The paper tackles the discrepancy between training and inference in neural machine translation due to intrinsic uncertainty, proposing uncertainty-aware semantic augmentation to capture universal semantic information among equivalent sentences, which significantly outperforms strong baselines and existing methods in experiments.
As a sequence-to-sequence generation task, neural machine translation (NMT) naturally contains intrinsic uncertainty, where a single sentence in one language has multiple valid counterparts in the other. However, the dominant methods for NMT only observe one of them from the parallel corpora for the model training but have to deal with adequate variations under the same meaning at inference. This leads to a discrepancy of the data distribution between the training and the inference phases. To address this problem, we propose uncertainty-aware semantic augmentation, which explicitly captures the universal semantic information among multiple semantically-equivalent source sentences and enhances the hidden representations with this information for better translations. Extensive experiments on various translation tasks reveal that our approach significantly outperforms the strong baselines and the existing methods.