Do Transformer Modifications Transfer Across Implementations and Applications?
Sharan Narang, Hyung Won Chung, Yi Tay, Liam Fedus, Thibault Fevry, Michael Matena, Karishma Malkan, Noah Fiedel, Noam Shazeer, Zhenzhong Lan, Yanqi Zhou, Wei Li, Nan Ding, Jake Marcus, Adam Roberts, Colin Raffel
Abstract
The research community has proposed copious modifications to the Transformer architecture since it was introduced over three years ago, relatively few of which have seen widespread adoption. In this paper, we comprehensively evaluate many of these modifications in a shared experimental setting that covers most of the common uses of the Transformer in natural language processing. Surprisingly, we find that most modifications do not meaningfully improve performance. Furthermore, most of the Transformer variants we found beneficial were either developed in the same codebase that we used or are relatively minor changes. We conjecture that performance improvements may strongly depend on implementation details and correspondingly make some recommendations for improving the generality of experimental results.- Anthology ID:
- 2021.emnlp-main.465
- 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:
- 5758–5773
- Language:
- URL:
- https://aclanthology.org/2021.emnlp-main.465
- DOI:
- 10.18653/v1/2021.emnlp-main.465
- Cite (ACL):
- Sharan Narang, Hyung Won Chung, Yi Tay, Liam Fedus, Thibault Fevry, Michael Matena, Karishma Malkan, Noah Fiedel, Noam Shazeer, Zhenzhong Lan, Yanqi Zhou, Wei Li, Nan Ding, Jake Marcus, Adam Roberts, and Colin Raffel. 2021. Do Transformer Modifications Transfer Across Implementations and Applications?. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5758–5773, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Do Transformer Modifications Transfer Across Implementations and Applications? (Narang et al., EMNLP 2021)
- PDF:
- https://preview.aclanthology.org/ingest-bitext-workshop/2021.emnlp-main.465.pdf
- Code
- google-research/google-research
- Data
- C4, WMT 2014, WebQuestions