15 papers
What Does It Mean to Forget a Person? Individual-Level Unlearning in Vision-Language Models
Xiongtao Sun, Hui Li, Tiantong Wu +3
Erasing individual identities from Vision-Language Models (VLMs) is uniquely challenging because personal data is entangled across modalities rather than stored as isolated attribu…
Concept Score Relearning: A Unified Cross-Architecture Attack on Concept Erasure
Hong Xi Tae, Jiaming Zhang, Wenwen He +2
Concept erasure aims to suppress undesirable knowledge in text-to-image generative models. However, existing robustness evaluations typically rely on relearning attacks tailored to…
Relevance Does Not Imply Applicability: Experience Activation for Personal GUI Agents
Fuyao Zhang, Xuan Wang, Zherui Li +3
Personal Graphical User Interface (GUI) agents rely on interaction history to infer what a user wants from ambiguous instructions and to anticipate recurring routines. Existing app…
SynChain: Inducing Computer-Use Agent Systems to Construct Their Own Attack Chains
Fuyao Zhang, Jiaming Zhang, Che Wang +7
Computer-use agents~(CUAs) have transformed large language models into persistent execution systems capable of generating, storing, and reusing artifacts like skills and memory ent…
XDomainBench: Diagnosing Reasoning Collapse in High-Dimensional Scientific Knowledge Composition
Gong Zhiren, Tiantong Wu, Jiaming Zhang +9
Large Language Models (LLMs) are increasingly deployed for knowledge synthesis, yet their capacity for compositional generalization in scientific knowledge remains under-characteri…
Position: Assistive Agents Need Accessibility Alignment
Jie Hu, Changyuan Yan, Yu Zheng +2
Assistive agents for Blind and Visually Impaired (BVI) users require accessibility alignment as a first-class design objective. Despite rapid progress in agentic AI, most systems a…
FraudBench: A Multimodal Benchmark for Detecting AI-Generated Fraudulent Refund Evidence
Xinyu Yan, Boyang Chen, Jiaming Zhang +12
Artificial Intelligence (AI)-generated images have become increasingly realistic and readily adaptable to real-world claims, creating new challenges for verifying visual evidence.…
ProjLens: Unveiling the Role of Projectors in Multimodal Model Safety
Kun Wang, Cheng Qian, Miao Yu +6
Multimodal Large Language Models (MLLMs) have achieved remarkable success in cross-modal understanding and generation, yet their deployment is threatened by critical safety vulnera…
AdapTools: Adaptive Tool-based Indirect Prompt Injection Attacks on Agentic LLMs
Che Wang, Jiaming Zhang, Ziqi Zhang +6
The integration of external data services (e.g., Model Context Protocol, MCP) has made large language model-based agents increasingly powerful for complex task execution. However,…
ICON: Indirect Prompt Injection Defense for Agents based on Inference-Time Correction
Che Wang, Fuyao Zhang, Jiaming Zhang +6
Large Language Model (LLM) agents are susceptible to Indirect Prompt Injection (IPI) attacks, where malicious instructions in retrieved content hijack the agent's execution. Existi…
VENOMREC: Cross-Modal Interactive Poisoning for Targeted Promotion in Multimodal LLM Recommender Systems
Guowei Guan, Yurong Hao, Jiaming Zhang +6
Multimodal large language models (MLLMs) are pushing recommender systems (RecSys) toward content-grounded retrieval and ranking via cross-modal fusion. We find that while cross-mod…
Disrupting Hierarchical Reasoning: Adversarial Protection for Geographic Privacy in Multimodal Reasoning Models
Jiaming Zhang, Che Wang, Yang Cao +2
Multi-modal large reasoning models (MLRMs) pose significant privacy risks by inferring precise geographic locations from personal images through hierarchical chain-of-thought reaso…
DualTAP: A Dual-Task Adversarial Protector for Mobile MLLM Agents
Fuyao Zhang, Jiaming Zhang, Che Wang +6
The reliance of mobile GUI agents on Multimodal Large Language Models (MLLMs) introduces a severe privacy vulnerability: screenshots containing Personally Identifiable Information…
MultiPriv: Benchmarking Individual-Level Privacy Reasoning in Vision-Language Models
Xiongtao Sun, Hui Li, Jiaming Zhang +5
Modern Vision-Language Models (VLMs) pose significant individual-level privacy risks by linking fragmented multimodal data to identifiable individuals through hierarchical chain-of…
CAVALRY-V: A Large-Scale Generator Framework for Adversarial Attacks on Video MLLMs
Jiaming Zhang, Rui Hu, Qing Guo +1
Video Multimodal Large Language Models (V-MLLMs) have shown impressive capabilities in temporal reasoning and cross-modal understanding, yet their vulnerability to adversarial atta…