Beyond BLEU: Training Neural Machine Translation with Semantic Similarity

TL;DR

This paper proposes a new semantic similarity-based reward function for training neural machine translation systems, which improves translation quality, convergence speed, and semantic diversity over traditional BLEU-based training.

Contribution

It introduces a novel semantic similarity metric for NMT training, addressing BLEU's limitations and enhancing translation performance and optimization efficiency.

Findings

Better translation quality as per BLEU, semantic similarity, and human evaluation.

Faster convergence during training.

Increased diversity and partial credit in scoring.

Abstract

While most neural machine translation (NMT) systems are still trained using maximum likelihood estimation, recent work has demonstrated that optimizing systems to directly improve evaluation metrics such as BLEU can substantially improve final translation accuracy. However, training with BLEU has some limitations: it doesn't assign partial credit, it has a limited range of output values, and it can penalize semantically correct hypotheses if they differ lexically from the reference. In this paper, we introduce an alternative reward function for optimizing NMT systems that is based on recent work in semantic similarity. We evaluate on four disparate languages translated to English, and find that training with our proposed metric results in better translations as evaluated by BLEU, semantic similarity, and human evaluation, and also that the optimization procedure converges faster. Analysis suggests that this is because the proposed metric is more conducive to optimization, assigning partial credit and providing more diversity in scores than BLEU.