9 papers
Mission (Im)Possible: Towards Autonomous Discovery of Autopilot Logic Flaws in Open-Source Drone Frameworks
Ziyi Liang, Zichuan Li, Bin Yuan +2
Autonomous drones are increasingly deployed in safety-critical tasks, yet their autopilot software can still contain subtle logic flaws that lead to unsafe behaviors. These flaws a…
Why Static Skill Inconsistencies Do Not Equate to Runtime Risks: A Behavioral Analysis of LLM Agent Execution
Fengrui Fan, Huahao Wang, Jingyi Fang +3
LLM agent frameworks heavily rely on external skill assets to standardize tool usage, encode operational workflows, and supply executable runtime resources. Despite their widesprea…
AGATE: Provenance-Based Runtime Defense Against Compositional Attacks on LLM Agents
Xiaorui Zhang, Zhuoran Cheng, Kailin Liu +6
LLM agents can produce harmful effects through sequences of ordinary operations. Judging such actions requires establishing both the authority that permits them and the origin of t…
MemArbiter: Decision-Time Memory Arbitration for Long-Horizon LLM Agents
Jiajun Dong, Yutao Hu, Fengrui Fan +3
Large language model (LLM) agents must retain and use cross-step information to act coherently in long-horizon tasks. Existing methods improve memory accessibility, yet action-rele…
Bulkhead: Automated Semantic Detection and Remediation of Container Escape Vulnerabilities
Qiyuan Fan, Zhi Li, Junjie Li +3
Container filesystem isolation depends on safely resolving container-controlled paths during privileged host-container interactions. As third-party plugins and agent sandboxes intr…
Towards Security-Auditable LLM Agents: A Unified Graph Representation
Chaofan Li, Lyuye Zhang, Jintao Zhai +9
LLM-based agentic systems are rapidly evolving to perform complex autonomous tasks through dynamic tool invocation, stateful memory management, and multi-agent collaboration. Howev…
A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook
Kaiwen Luo, Zhenhong Zhou, Leyan Wang +32
Advances in Large Language Models (LLMs) have paved the way for Multimodal Large Language Models (MLLMs). Among these, Large Audio Language Models (LALMs) are essential for realizi…
SoK: Automated Vulnerability Repair: Methods, Tools, and Assessments
Yiwei Hu, Zhen Li, Kedie Shu +5
The increasing complexity of software has led to the steady growth of vulnerabilities. Vulnerability repair investigates how to fix software vulnerabilities. Manual vulnerability r…
On the Effectiveness of Function-Level Vulnerability Detectors for Inter-Procedural Vulnerabilities
Zhen Li, Ning Wang, Deqing Zou +5
Software vulnerabilities are a major cyber threat and it is important to detect them. One important approach to detecting vulnerabilities is to use deep learning while treating a p…