JailBreakV: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks
Weidi Luo, Siyuan Ma, Xiaogeng Liu, Guo, Xiaoyu, Chaowei Xiao
De quem é
- University of Wisconsin–Madisonafiliação declarada pelos autores, segundo o OpenAlex
Outros papers das mesmas organizações
- Voyager: An Open-Ended Embodied Agent with Large Language ModelsThe University of Texas at Austin · 3.453 citações
- Generalized Out-of-Distribution Detection: A SurveyNanyang Technological University · 1.775 citações
- Interpretable Preferences via Multi-Objective Reward Modeling and Mixture-of-ExpertsUniversity of Illinois Urbana-Champaign · 472 citações
- Understanding Large-Language Model (LLM)-powered Human-Robot InteractionUniversity of Wisconsin–Madison · 335 citações
- LLM Inference Unveiled: Survey and Roofline Model InsightsUniversity of Wisconsin–Madison · 316 citações
- LLaVA-Plus: Learning to Use Tools for Creating Multimodal AgentsUniversity of Wisconsin–Madison · 305 citações