13.06.2025
Volker Tresp and His Team Win ReGenAI @CVPR 2025 Best Paper Award
Honored for Work on New Jailbreak Vulnerability in T2I Diffusion Models
MCML Junior Members Tong Liu, Gengyuan Zhang, Shuo Chen, and MCML PI Volker Tresp and their co-authors have received the Best Paper Award at the Second Workshop on Responsible Generative AI (ReGenAI) Workshop at CVPR 2025 for their paper “Multimodal Pragmatic Jailbreak on Text-to-image Models”.
The authors show that text-to-image models can be easily exploited to produce unsafe content through cross‑modal interactions between safe text and images, a vulnerability that current safety filters fail to address effectively.
Congratulations from us!
Check out the full paper:
Multimodal Pragmatic Jailbreak on Text-to-image Models.
ACL 2025 - 63rd Annual Meeting of the Association for Computational Linguistics. Vienna, Austria, Jul 27-Aug 01, 2025. DOI GitHub
Related
18.12.2025
"See, Don’t Assume": Revealing and Reducing Gender Bias in AI
ICLR 2025 research led by Zeynep Akata’s team reveals and reduces gender bias in popular vision-language AI models.
16.12.2025
Fabian Theis Featured in Handelsblatt on the Future of AI in Precision Medicine
MCML PI Fabian Theis discusses AI-driven precision medicine and its growing impact on individualized healthcare and biomedical research.
16.12.2025
Gitta Kutyniok Featured in VDI Nachrichten on AI Ethics
Gitta Kutyniok discusses measurable criteria for ethical AI, promoting safe and responsible autonomous decision-making.
16.12.2025
Hinrich Schütze Featured in WirtschaftsWoche on Innovative AI Approaches
Hinrich Schütze discusses Giotto.ai’s efficient AI models, highlighting memory separation and context-aware decoding to improve robustness.
