works on

From the 1 of 9 linked papers with an AI index.

activity
20242026
collaborators

9 papers

cs.CR2026

When Agents Learn to Be You: Benchmarking Privacy Leakage, Impersonation Risk, and Defenses in Persona Skills

Yongli Xiang, Zhifang Zhang, Bojun Yang +4

Persona skills distill personal interaction histories into portable and executable artifacts for downstream agents. While enabling flexible personalization, this process concentrat…

cs.CV2026

TokenSwap: Backdoor Attack on the Compositional Understanding of Large Vision-Language Models

Zhifang Zhang, Qiqi Tao, Jiaqi Lv +3

The paper introduces TokenSwap, a stealthy backdoor attack on large vision-language models that swaps key textual tokens to corrupt the model's understanding of object relationship…

cs.CR2026

Poise: Position-Aware One-Instruction Skill Injection for Silent Execution on LLM Agents

Haochang Hao, Dehai Min, Zhifang Zhang +4

Agent skills extend general-purpose agents, but their open format enables skill poisoning: a tampered skill can make an agent run an attacker's command while completing the user's…

cs.AI2026

Towards Safer Large Reasoning Models by Promoting Safety Decision-Making before Chain-of-Thought Generation

Jianan Chen, Zhifang Zhang, Shuo He +3

Large reasoning models (LRMs) achieved remarkable performance via chain-of-thought (CoT), but recent studies showed that such enhanced reasoning capabilities are at the expense of…

cs.CV2026

Test-Time Attention Purification for Backdoored Large Vision Language Models

Zhifang Zhang, Bojun Yang, Shuo He +5

Despite the strong multimodal performance, large vision-language models (LVLMs) are vulnerable during fine-tuning to backdoor attacks, where adversaries insert trigger-embedded sam…

cs.CV2025

Defending Multimodal Backdoored Models by Repulsive Visual Prompt Tuning

Zhifang Zhang, Shuo He, Haobo Wang +2

Multimodal contrastive learning models (e.g., CLIP) can learn high-quality representations from large-scale image-text datasets, while they exhibit significant vulnerabilities to b…