Treffer: Con Instruction: Universal Jailbreaking of Multimodal Large Language Models via Non-Textual Modalities

Titel:
Veröffentlicht in:
Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics. Volume 1 Long Papers. - Band 1, Heft /p2917–2933 (2025)
Entstehung:
2025
Format:
E-Artikel
Sprache:
Englisch
DOI:
10.18653/v1/2025.acl-long.146
Open Access Rechte:
Open Access
CC BY 4.0