Agreement-Constrained Probabilistic Minimum Bayes Risk Decoding

Koki Natsumi, Hiroyuki Deguchi, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe


Abstract
Minimum Bayes risk (MBR) decoding generates high-quality translations by maximizing the expected utility of output candidates, but it evaluates all pairwise scores over the candidate set; hence, it takes quadratic time with respect to the number of candidates. To reduce the number of utility function calls, probabilistic MBR (PMBR) decoding partially evaluates quality scores using sampled pairs of candidates and completes the missing scores with a matrix completion algorithm. Nevertheless, it degrades the translation quality as the number of utility function calls is reduced. Therefore, to improve the trade-off between quality and cost, we propose agreement-constrained PMBR (AC-PMBR) decoding, which leverages a knowledge distilled model to guide the completion of the score matrix. Our AC-PMBR decoding improved approximation errors of matrix completion by up to 3 times and achieved higher translation quality compared with PMBR decoding at a comparable computational cost on the WMT’23 En↔De translation tasks.
Anthology ID:
2025.ijcnlp-short.39
Volume:
Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics
Month:
December
Year:
2025
Address:
Mumbai, India
Editors:
Kentaro Inui, Sakriani Sakti, Haofen Wang, Derek F. Wong, Pushpak Bhattacharyya, Biplab Banerjee, Asif Ekbal, Tanmoy Chakraborty, Dhirendra Pratap Singh
Venues:
IJCNLP | AACL
SIG:
Publisher:
The Asian Federation of Natural Language Processing and The Association for Computational Linguistics
Note:
Pages:
484–493
Language:
URL:
https://preview.aclanthology.org/ingest-ijcnlp-aacl/2025.ijcnlp-short.39/
DOI:
Bibkey:
Cite (ACL):
Koki Natsumi, Hiroyuki Deguchi, Yusuke Sakai, Hidetaka Kamigaito, and Taro Watanabe. 2025. Agreement-Constrained Probabilistic Minimum Bayes Risk Decoding. In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, pages 484–493, Mumbai, India. The Asian Federation of Natural Language Processing and The Association for Computational Linguistics.
Cite (Informal):
Agreement-Constrained Probabilistic Minimum Bayes Risk Decoding (Natsumi et al., IJCNLP-AACL 2025)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-ijcnlp-aacl/2025.ijcnlp-short.39.pdf