15 citations · 15 across the 2 of their papers we have counts for
5 papers
Understanding and Mitigating Token-Pruning-Induced Vulnerabilities in VLMs
Shuailong Wang, Xinyu Lyu, Shengming Yuan +3
Token-Pruning accelerates Vision-Language Models by removing redundant visual tokens, yet its safety implications remain underexplored. In this work, we present the first comprehen…
FlexAC: Towards Flexible Control of Associative Reasoning in Multimodal Large Language Models
Shengming Yuan, Xinyu Lyu, Shuailong Wang +3
Multimodal large language models (MLLMs) face an inherent trade-off between faithfulness and creativity, as different tasks require varying degrees of associative reasoning. Howeve…
SafePTR: Token-Level Jailbreak Defense in Multimodal LLMs via Prune-then-Restore Mechanism
Beitao Chen, Xinyu Lyu, Lianli Gao +2
By incorporating visual inputs, Multimodal Large Language Models (MLLMs) extend LLMs to support visual reasoning. However, this integration also introduces new vulnerabilities, mak…
Any Target Can be Offense: Adversarial Example Generation via Generalized Latent Infection
Youheng Sun, Shengming Yuan, Xuanhan Wang +2
Targeted adversarial attack, which aims to mislead a model to recognize any image as a target object by imperceptible perturbations, has become a mainstream tool for vulnerability…
Natural Color Fool: Towards Boosting Black-box Unrestricted Attacks
Shengming Yuan, Qilong Zhang, Lianli Gao +2
Unrestricted color attacks, which manipulate semantically meaningful color of an image, have shown their stealthiness and success in fooling both human eyes and deep neural network…