Multimodal Pretraining Unmasked: A Meta-Analysis and a Unified Framework of Vision-and-Language BERTs
Emanuele Bugliarello, Ryan Cotterell, Naoaki Okazaki, Desmond Elliott
Abstract
Large-scale pretraining and task-specific fine- tuning is now the standard methodology for many tasks in computer vision and natural language processing. Recently, a multitude of methods have been proposed for pretraining vision and language BERTs to tackle challenges at the intersection of these two key areas of AI. These models can be categorized into either single-stream or dual-stream encoders. We study the differences between these two categories, and show how they can be unified under a single theoretical framework. We then conduct controlled experiments to discern the empirical differences between five vision and language BERTs. Our experiments show that training data and hyperparameters are responsible for most of the differences between the reported results, but they also reveal that the embedding layer plays a crucial role in these massive models.- Anthology ID:
- 2021.tacl-1.58
- Volume:
- Transactions of the Association for Computational Linguistics, Volume 9
- Month:
- Year:
- 2021
- Address:
- Cambridge, MA
- Editors:
- Brian Roark, Ani Nenkova
- Venue:
- TACL
- SIG:
- Publisher:
- MIT Press
- Note:
- Pages:
- 978–994
- Language:
- URL:
- https://aclanthology.org/2021.tacl-1.58
- DOI:
- 10.1162/tacl_a_00408
- Cite (ACL):
- Emanuele Bugliarello, Ryan Cotterell, Naoaki Okazaki, and Desmond Elliott. 2021. Multimodal Pretraining Unmasked: A Meta-Analysis and a Unified Framework of Vision-and-Language BERTs. Transactions of the Association for Computational Linguistics, 9:978–994.
- Cite (Informal):
- Multimodal Pretraining Unmasked: A Meta-Analysis and a Unified Framework of Vision-and-Language BERTs (Bugliarello et al., TACL 2021)
- PDF:
- https://preview.aclanthology.org/naacl-24-ws-corrections/2021.tacl-1.58.pdf