Abstract
Nearest neighbor machine translation (kNN-MT), which interpolates target token probabilities with estimates derived from additional examples, has achieved significant improvements and attracted extensive interest in recent years. However, existing research does not explicitly consider the source context when retrieving similar examples, potentially leading to suboptimal performance. To address this, we comprehensively revisit the role of source context and propose a simple and effective method for improving neural machine translation via source context enhancement, demonstrating its crucial role in both retrieving superior examples and determining more suitable interpolation coefficients. Furthermore, we reveal that the probability estimation can be further optimized by incorporating a source-aware distance calibration module. Comprehensive experiments show that our proposed approach can be seamlessly integrated with representative kNN-MT baselines, resulting in substantial improvements over these strong baselines across a number of settings and domains. Remarkably, these improvements can reach up to 1.6 BLEU points.- Anthology ID:
- 2023.emnlp-main.503
- Volume:
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 8087–8098
- Language:
- URL:
- https://aclanthology.org/2023.emnlp-main.503
- DOI:
- 10.18653/v1/2023.emnlp-main.503
- Cite (ACL):
- Xuanhong Li, Peng Li, and Po Hu. 2023. Revisiting Source Context in Nearest Neighbor Machine Translation. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 8087–8098, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Revisiting Source Context in Nearest Neighbor Machine Translation (Li et al., EMNLP 2023)
- PDF:
- https://preview.aclanthology.org/corrections-2024-05/2023.emnlp-main.503.pdf