Aishan Liu
2 indexed papers
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
226
Top categories
Crypto×2Vision×1
Frequent co-authors
Research Timeline
2026
AgentVisor: Defending LLM Agents Against Prompt Injection via Semantic Virtualization
AgentVisor is a novel defense framework that uses semantic virtualization, inspired by OS principles, to significantly reduce LLM agent vulnerability to prompt injection while maintaining high utility.
Seeing Is Not Screening: Multimodal Hidden Instruction Attacks on Agent Skill Scanners
SkillCamo: A document-mediated multimodal instruction attack
Highlighted terms show continued research focus across papers
Papers
cs.CRcs.CVEmpiricalRecentJun 16, 2026
Seeing Is Not Screening: Multimodal Hidden Instruction Attacks on Agent Skill Scanners
Xiaojun Jia, Jie Liao, Simeng Qin, Ke Ma +4 more
SkillCamo: A document-mediated multimodal instruction attack
View →
cs.CRApr 27, 2026