Asymmetric Bias in Text-to-Image Generation with Adversarial Attacks

Haz Shahgir, Xianghao Kong, Greg Ver Steeg, Yue Dong


Abstract
The widespread use of Text-to-Image (T2I) models in content generation requires careful examination of their safety, including their robustness to adversarial attacks. Despite extensive research on adversarial attacks, the reasons for their effectiveness remain underexplored. This paper presents an empirical study on adversarial attacks against T2I models, focusing on analyzing factors associated with attack success rates (ASR). We introduce a new attack objective - entity swapping using adversarial suffixes and two gradient-based attack algorithms. Human and automatic evaluations reveal the asymmetric nature of ASRs on entity swap: for example, it is easier to replace “human” with “robot” in the prompt “a human dancing in the rain.” with an adversarial suffix, but the reverse replacement is significantly harder. We further propose probing metrics to establish indicative signals from the model’s beliefs to the adversarial ASR. We identify conditions that result in a success probability of 60% for adversarial attacks and others where this likelihood drops below 5%. The code and data are available at https://github.com/Patchwork53/AsymmetricAttack
Anthology ID:
2024.findings-acl.344
Volume:
Findings of the Association for Computational Linguistics: ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5779–5796
Language:
URL:
https://aclanthology.org/2024.findings-acl.344
DOI:
10.18653/v1/2024.findings-acl.344
Bibkey:
Cite (ACL):
Haz Shahgir, Xianghao Kong, Greg Ver Steeg, and Yue Dong. 2024. Asymmetric Bias in Text-to-Image Generation with Adversarial Attacks. In Findings of the Association for Computational Linguistics: ACL 2024, pages 5779–5796, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Asymmetric Bias in Text-to-Image Generation with Adversarial Attacks (Shahgir et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-acl.344.pdf