Abstract
Transformer becomes the state-of-the-art translation model, while it is not well studied how each intermediate component contributes to the model performance, which poses significant challenges for designing optimal architectures. In this work, we bridge this gap by evaluating the impact of individual component (sub-layer) in trained Transformer models from different perspectives. Experimental results across language pairs, training strategies, and model capacities show that certain components are consistently more important than the others. We also report a number of interesting findings that might help humans better analyze, understand and improve Transformer models. Based on these observations, we further propose a new training strategy that can improves translation performance by distinguishing the unimportant components in training.- Anthology ID:
- 2020.coling-main.529
- Volume:
- Proceedings of the 28th International Conference on Computational Linguistics
- Month:
- December
- Year:
- 2020
- Address:
- Barcelona, Spain (Online)
- Editors:
- Donia Scott, Nuria Bel, Chengqing Zong
- Venue:
- COLING
- SIG:
- Publisher:
- International Committee on Computational Linguistics
- Note:
- Pages:
- 6019–6029
- Language:
- URL:
- https://aclanthology.org/2020.coling-main.529
- DOI:
- 10.18653/v1/2020.coling-main.529
- Cite (ACL):
- Wenxuan Wang and Zhaopeng Tu. 2020. Rethinking the Value of Transformer Components. In Proceedings of the 28th International Conference on Computational Linguistics, pages 6019–6029, Barcelona, Spain (Online). International Committee on Computational Linguistics.
- Cite (Informal):
- Rethinking the Value of Transformer Components (Wang & Tu, COLING 2020)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2020.coling-main.529.pdf