6 citations · 18 across the 41 of their papers we have counts for
5 papers · 1 filter
HarnessRisk: A Lifecycle-Oriented Benchmark for Agent Harness Safety
Yajing Bai, Jinhao Duan, Jie Peng +4
Large language models are increasingly deployed through agent harnesses that manage tools, extensions, persistent state, permissions, and external actions. Existing safety benchmar…
Dynamic Malicious Skills in Agentic AI
Tianhao Chen, Zhengyuan Jiang, Yuepeng Hu +2
Skills are a key enabling component of agentic AI. While they enhance agents' capabilities, they also introduce new attack surfaces. In this work, we investigate one such attack su…
Measuring Real-World Prompt Injection Attacks in LLM-based Resume Screening
Mohan Zhang, Yuqi Jia, Zhen Tan +4
LLMs are vulnerable to prompt injection attacks. However, this vulnerability has been primarily demonstrated conceptually in academic studies or through a few anecdotal case studie…
"Glue pizza and eat rocks" -- Exploiting Vulnerabilities in Retrieval-Augmented Generative Models
Zhen Tan, Chengshuai Zhao, Raha Moraffah +5
Retrieval-Augmented Generative (RAG) models enhance Large Language Models (LLMs) by integrating external knowledge bases, improving their performance in applications like fact-chec…
The Wolf Within: Covert Injection of Malice into MLLM Societies via an MLLM Operative
Zhen Tan, Chengshuai Zhao, Raha Moraffah +4
Due to their unprecedented ability to process and respond to various types of data, Multimodal Large Language Models (MLLMs) are constantly defining the new boundary of Artificial…