Focus on the Target’s Vocabulary: Masked Label Smoothing for Machine Translation

Liang Chen, Runxin Xu, Baobao Chang


Abstract
Label smoothing and vocabulary sharing are two widely used techniques in neural machine translation models. However, we argue that simply applying both techniques can be conflicting and even leads to sub-optimal performance. When allocating smoothed probability, original label smoothing treats the source-side words that would never appear in the target language equally to the real target-side words, which could bias the translation model. To address this issue, we propose Masked Label Smoothing (MLS), a new mechanism that masks the soft label probability of source-side words to zero. Simple yet effective, MLS manages to better integrate label smoothing with vocabulary sharing. Our extensive experiments show that MLS consistently yields improvement over original label smoothing on different datasets, including bilingual and multilingual translation from both translation quality and model’s calibration. Our code is released at https://github.com/PKUnlp-icler/MLS
Anthology ID:
2022.acl-short.74
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
665–671
Language:
URL:
https://aclanthology.org/2022.acl-short.74
DOI:
10.18653/v1/2022.acl-short.74
Bibkey:
Cite (ACL):
Liang Chen, Runxin Xu, and Baobao Chang. 2022. Focus on the Target’s Vocabulary: Masked Label Smoothing for Machine Translation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 665–671, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Focus on the Target’s Vocabulary: Masked Label Smoothing for Machine Translation (Chen et al., ACL 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/add_acl24_videos/2022.acl-short.74.pdf
Software:
 2022.acl-short.74.software.zip
Code
 chenllliang/MLS +  additional community code