How does the task complexity of masked pretraining objectives affect downstream performance?

Atsuki Yamaguchi, Hiroaki Ozaki, Terufumi Morishita, Gaku Morio, Yasuhiro Sogawa


Abstract
Masked language modeling (MLM) is a widely used self-supervised pretraining objective, where a model needs to predict an original token that is replaced with a mask given contexts. Although simpler and computationally efficient pretraining objectives, e.g., predicting the first character of a masked token, have recently shown comparable results to MLM, no objectives with a masking scheme actually outperform it in downstream tasks. Motivated by the assumption that their lack of complexity plays a vital role in the degradation, we validate whether more complex masked objectives can achieve better results and investigate how much complexity they should have to perform comparably to MLM. Our results using GLUE, SQuAD, and Universal Dependencies benchmarks demonstrate that more complicated objectives tend to show better downstream results with at least half of the MLM complexity needed to perform comparably to MLM. Finally, we discuss how we should pretrain a model using a masked objective from the task complexity perspective.
Anthology ID:
2023.findings-acl.669
Volume:
Findings of the Association for Computational Linguistics: ACL 2023
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
10527–10537
Language:
URL:
https://aclanthology.org/2023.findings-acl.669
DOI:
10.18653/v1/2023.findings-acl.669
Bibkey:
Cite (ACL):
Atsuki Yamaguchi, Hiroaki Ozaki, Terufumi Morishita, Gaku Morio, and Yasuhiro Sogawa. 2023. How does the task complexity of masked pretraining objectives affect downstream performance?. In Findings of the Association for Computational Linguistics: ACL 2023, pages 10527–10537, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
How does the task complexity of masked pretraining objectives affect downstream performance? (Yamaguchi et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.669.pdf
Video:
 https://aclanthology.org/2023.findings-acl.669.mp4