Unveiling the Power of Source: Source-based Minimum Bayes Risk Decoding for Neural Machine Translation

2024/06/17 by Boxuan Lyu, Hidetaka Kamigaito, Lyu, Boxuan +5
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques

paper · pdf · doi:10.48550/arxiv.2406.11632

openalex publication_date 2024/06/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Maximum a posteriori decoding, a commonly used method for neural machine translation (NMT), aims to maximize the estimated posterior probability. However, high estimated probability does not always lead to high translation quality. Minimum Bayes Risk (MBR) decoding offers an alternative by seeking hypotheses with the highest expected utility. Inspired by Quality Estimation (QE) reranking which uses the QE model as a ranker we propose source-based MBR (sMBR) decoding, a novel approach that utilizes quasi-sources (generated via paraphrasing or back-translation) as ``support hypotheses'' and a reference-free quality estimation metric as the utility function, marking the first work to solely use sources in MBR decoding. Experiments show that sMBR outperforms QE reranking and the standard MBR decoding. Our findings suggest that sMBR is a promising approach for NMT decoding.

Related

• High Quality Rather than High Model Probability: Minimum Bayes Risk Decoding with Neural Metrics
• Quality-Aware Translation Models: Efficient Generation and Quality Estimation in a Single Model
• Understanding the Properties of Minimum Bayes Risk Decoding in Neural Machine Translation
• Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation
• Quality-Aware Decoding for Neural Machine Translation