CARETS: A Consistency And Robustness Evaluative Test Suite for VQA

Carlos E. Jimenez, Olga Russakovsky, Karthik Narasimhan


Abstract
We introduce CARETS, a systematic test suite to measure consistency and robustness of modern VQA models through a series of six fine-grained capability tests. In contrast to existing VQA test sets, CARETS features balanced question generation to create pairs of instances to test models, with each pair focusing on a specific capability such as rephrasing, logical symmetry or image obfuscation. We evaluate six modern VQA systems on CARETS and identify several actionable weaknesses in model comprehension, especially with concepts such as negation, disjunction, or hypernym invariance. Interestingly, even the most sophisticated models are sensitive to aspects such as swapping the order of terms in a conjunction or varying the number of answer choices mentioned in the question. We release CARETS to be used as an extensible tool for evaluating multi-modal model robustness.
Anthology ID:
2022.acl-long.443
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6392–6405
Language:
URL:
https://aclanthology.org/2022.acl-long.443
DOI:
10.18653/v1/2022.acl-long.443
Bibkey:
Cite (ACL):
Carlos E. Jimenez, Olga Russakovsky, and Karthik Narasimhan. 2022. CARETS: A Consistency And Robustness Evaluative Test Suite for VQA. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6392–6405, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
CARETS: A Consistency And Robustness Evaluative Test Suite for VQA (Jimenez et al., ACL 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2022.acl-long.443.pdf
Software:
 2022.acl-long.443.software.tgz
Video:
 https://preview.aclanthology.org/ingestion-script-update/2022.acl-long.443.mp4
Code
 princeton-nlp/carets
Data
GQAVisual GenomeVisual Question Answering