Qi, Huiyan

Benchmarking Gaslighting Negation Attacks Against Multimodal Large Language Models

2025 · arXiv (Cornell University)

Multimodal Large Language Models (MLLMs) have exhibited remarkable advancements in integrating different modalities, excelling in complex understanding and generation tasks. Despite their success, MLLMs remain vulnerable to conversational adversarial inputs. In this paper, we systematically …