SupCL-Seq: Supervised Contrastive Learning for Downstream Optimized Sequence Representations

Hooman Sedghamiz, Shivam Raval, Enrico Santus, Tuka Alhanai, Mohammad Ghassemi


Abstract
While contrastive learning is proven to be an effective training strategy in computer vision, Natural Language Processing (NLP) is only recently adopting it as a self-supervised alternative to Masked Language Modeling (MLM) for improving sequence representations. This paper introduces SupCL-Seq, which extends the supervised contrastive learning from computer vision to the optimization of sequence representations in NLP. By altering the dropout mask probability in standard Transformer architectures (e.g. BERT-base), for every representation (anchor), we generate augmented altered views. A supervised contrastive loss is then utilized to maximize the system’s capability of pulling together similar samples (e.g., anchors and their altered views) and pushing apart the samples belonging to the other classes. Despite its simplicity, SupCL-Seq leads to large gains in many sequence classification tasks on the GLUE benchmark compared to a standard BERT-base, including 6% absolute improvement on CoLA, 5.4% on MRPC, 4.7% on RTE and 2.6% on STS-B. We also show consistent gains over self-supervised contrastively learned representations, especially in non-semantic tasks. Finally we show that these gains are not solely due to augmentation, but rather to a downstream optimized sequence representation.
Anthology ID:
2021.findings-emnlp.289
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2021
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
Findings
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
3398–3403
Language:
URL:
https://aclanthology.org/2021.findings-emnlp.289
DOI:
10.18653/v1/2021.findings-emnlp.289
Bibkey:
Cite (ACL):
Hooman Sedghamiz, Shivam Raval, Enrico Santus, Tuka Alhanai, and Mohammad Ghassemi. 2021. SupCL-Seq: Supervised Contrastive Learning for Downstream Optimized Sequence Representations. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 3398–3403, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
SupCL-Seq: Supervised Contrastive Learning for Downstream Optimized Sequence Representations (Sedghamiz et al., Findings 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.findings-emnlp.289.pdf
Video:
 https://aclanthology.org/2021.findings-emnlp.289.mp4
Code
 hooman650/supcl-seq
Data
CoLAGLUE