Abstract
We study the power of cross-attention in the Transformer architecture within the context of transfer learning for machine translation, and extend the findings of studies into cross-attention when training from scratch. We conduct a series of experiments through fine-tuning a translation model on data where either the source or target language has changed. These experiments reveal that fine-tuning only the cross-attention parameters is nearly as effective as fine-tuning all parameters (i.e., the entire translation model). We provide insights into why this is the case and observe that limiting fine-tuning in this manner yields cross-lingually aligned embeddings. The implications of this finding for researchers and practitioners include a mitigation of catastrophic forgetting, the potential for zero-shot translation, and the ability to extend machine translation models to several new language pairs with reduced parameter storage overhead.- Anthology ID:
- 2021.emnlp-main.132
- Volume:
- Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
- Month:
- November
- Year:
- 2021
- Address:
- Online and Punta Cana, Dominican Republic
- Editors:
- Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1754–1765
- Language:
- URL:
- https://aclanthology.org/2021.emnlp-main.132
- DOI:
- 10.18653/v1/2021.emnlp-main.132
- Cite (ACL):
- Mozhdeh Gheini, Xiang Ren, and Jonathan May. 2021. Cross-Attention is All You Need: Adapting Pretrained Transformers for Machine Translation. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 1754–1765, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Cross-Attention is All You Need: Adapting Pretrained Transformers for Machine Translation (Gheini et al., EMNLP 2021)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2021.emnlp-main.132.pdf
- Code
- mgheini/xattn-transfer-for-mt