Modular Visual Question Answering via Code Generation

Sanjay Subramanian, Medhini Narasimhan, Kushal Khangaonkar, Kevin Yang, Arsha Nagrani, Cordelia Schmid, Andy Zeng, Trevor Darrell, Dan Klein


Abstract
We present a framework that formulates visual question answering as modular code generation. In contrast to prior work on modular approaches to VQA, our approach requires no additional training and relies on pre-trained language models (LMs), visual models pre-trained on image-caption pairs, and fifty VQA examples used for in-context learning. The generated Python programs invoke and compose the outputs of the visual models using arithmetic and conditional logic. Our approach improves accuracy on the COVR dataset by at least 3% and on the GQA dataset by 2% compared to the few-shot baseline that does not employ code generation.
Anthology ID:
2023.acl-short.65
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
747–761
Language:
URL:
https://aclanthology.org/2023.acl-short.65
DOI:
10.18653/v1/2023.acl-short.65
Bibkey:
Cite (ACL):
Sanjay Subramanian, Medhini Narasimhan, Kushal Khangaonkar, Kevin Yang, Arsha Nagrani, Cordelia Schmid, Andy Zeng, Trevor Darrell, and Dan Klein. 2023. Modular Visual Question Answering via Code Generation. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 747–761, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Modular Visual Question Answering via Code Generation (Subramanian et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-short.65.pdf
Video:
 https://aclanthology.org/2023.acl-short.65.mp4