Don’t Judge a Language Model by Its Last Layer: Contrastive Learning with Layer-Wise Attention Pooling
Dongsuk Oh, Yejin Kim, Hodong Lee, H. Howie Huang, Heuiseok Lim
Abstract
Recent pre-trained language models (PLMs) achieved great success on many natural language processing tasks through learning linguistic features and contextualized sentence representation. Since attributes captured in stacked layers of PLMs are not clearly identified, straightforward approaches such as embedding the last layer are commonly preferred to derive sentence representations from PLMs. This paper introduces the attention-based pooling strategy, which enables the model to preserve layer-wise signals captured in each layer and learn digested linguistic features for downstream tasks. The contrastive learning objective can adapt the layer-wise attention pooling to both unsupervised and supervised manners. It results in regularizing the anisotropic space of pre-trained embeddings and being more uniform. We evaluate our model on standard semantic textual similarity (STS) and semantic search tasks. As a result, our method improved the performance of the base contrastive learned BERTbase and variants.- Anthology ID:
- 2022.coling-1.405
- Volume:
- Proceedings of the 29th International Conference on Computational Linguistics
- Month:
- October
- Year:
- 2022
- Address:
- Gyeongju, Republic of Korea
- Editors:
- Nicoletta Calzolari, Chu-Ren Huang, Hansaem Kim, James Pustejovsky, Leo Wanner, Key-Sun Choi, Pum-Mo Ryu, Hsin-Hsi Chen, Lucia Donatelli, Heng Ji, Sadao Kurohashi, Patrizia Paggio, Nianwen Xue, Seokhwan Kim, Younggyun Hahm, Zhong He, Tony Kyungil Lee, Enrico Santus, Francis Bond, Seung-Hoon Na
- Venue:
- COLING
- SIG:
- Publisher:
- International Committee on Computational Linguistics
- Note:
- Pages:
- 4585–4592
- Language:
- URL:
- https://aclanthology.org/2022.coling-1.405
- DOI:
- Cite (ACL):
- Dongsuk Oh, Yejin Kim, Hodong Lee, H. Howie Huang, and Heuiseok Lim. 2022. Don’t Judge a Language Model by Its Last Layer: Contrastive Learning with Layer-Wise Attention Pooling. In Proceedings of the 29th International Conference on Computational Linguistics, pages 4585–4592, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.
- Cite (Informal):
- Don’t Judge a Language Model by Its Last Layer: Contrastive Learning with Layer-Wise Attention Pooling (Oh et al., COLING 2022)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-3/2022.coling-1.405.pdf
- Code
- nlpods/layerattpooler