arXiv:2402.04251 · v3

Linear-time Minimum Bayes Risk Decoding with Reference Aggregation

首次提交 2024-02-07 02:59

版本更新 2026-11-10 16:05

雷达观测 2026-11-11 13:08

摘要

Minimum Bayes Risk (MBR) decoding is a text generation technique that has been shown to improve the quality of machine translations, but is expensive, even if a sampling-based approximation is used. Besides requiring a large number of sampled sequences, it requires the pairwise calculation of a utility metric, which has quadratic complexity. In this paper, we propose to approximate pairwise metric scores with scores calculated against aggregated reference representations. This changes the complexity of utility estimation from $O(n^2)$ to $O(n)$, while empirically preserving most of the quality gains of MBR decoding. We release our source code at https://github.com/ZurichNLP/mbr

作者备注：ACL 2024. v3: More accurate time measurements for ChrF