Quality-Aware Decoding for Neural Machine Translation
This work addresses a bottleneck in neural machine translation for practitioners by integrating quality estimation into decoding, though it is incremental as it builds on existing evaluation methods.
The paper tackles the problem that neural machine translation decoding typically ignores quality estimation, and shows that incorporating quality-aware decoding methods consistently outperforms standard maximum a posteriori decoding across multiple datasets and models, with gains in both automatic metrics and human assessments.
Despite the progress in machine translation quality estimation and evaluation in the last years, decoding in neural machine translation (NMT) is mostly oblivious to this and centers around finding the most probable translation according to the model (MAP decoding), approximated with beam search. In this paper, we bring together these two lines of research and propose quality-aware decoding for NMT, by leveraging recent breakthroughs in reference-free and reference-based MT evaluation through various inference methods like $N$-best reranking and minimum Bayes risk decoding. We perform an extensive comparison of various possible candidate generation and ranking methods across four datasets and two model classes and find that quality-aware decoding consistently outperforms MAP-based decoding according both to state-of-the-art automatic metrics (COMET and BLEURT) and to human assessments. Our code is available at https://github.com/deep-spin/qaware-decode.