8 papers
ClawSentry: A Progressive Multi-Tier Security Monitor for Safeguarding Autonomous LLM Agents
Kai Wang, Zeming Wei, BiaoJie Zeng +7
As large language model (LLM) agents move from conversation to executing code, reading local files, and orchestrating external tools, a single agent hijacked by a malicious third-p…
Privacy-Preserving Robustness Verification for Neural Networks
Nianyun Song, Xiaokun Luan, Yu Guo +3
Neural network verification and data privacy are inherently in tension: verification demands full access to model parameters and input data, yet both are increasingly restricted by…
VOW: Verifiable and Oblivious Watermark Detection for Large Language Models
Xiaokun Luan, Yihao Zhang, Pengcheng Su +4
Regulation requires generated text to be marked and detectable, but symmetric-key watermarks leave detection to the key-holding provider. We study first-party provenance audit (FPA…
AgentWorm: Self-Propagating Attacks Across LLM Agent Ecosystems
Yihao Zhang, Zeming Wei, Xiaokun Luan +7
Autonomous LLM-based agents increasingly operate as long-running processes forming densely interconnected multi-agent ecosystems, whose security properties remain largely unexplore…
RACC: Representation-Aware Coverage Criteria for LLM Safety Testing
Zeming Wei, Zhixin Zhang, Chengcan Wu +3
Large Language Models (LLMs) face severe safety risks from jailbreak attacks, yet current safety testing largely relies on static datasets and lacks systematic criteria to evaluate…
Beyond Correctness: Exposing LLM-generated Logical Flaws in Reasoning via Multi-step Automated Theorem Proving
Xinyi Zheng, Ningke Li, Xiaokun Luan +4
Large Language Models (LLMs) have demonstrated impressive reasoning capabilities, leading to their adoption in high-stakes domains such as healthcare, law, and scientific research.…
Automata-Based Steering of Large Language Models for Diverse Structured Generation
Xiaokun Luan, Zeming Wei, Yihao Zhang +1
Large language models (LLMs) are increasingly tasked with generating structured outputs. While structured generation methods ensure validity, they often lack output diversity, a cr…
Protecting Deep Learning Model Copyrights with Adversarial Example-Free Reuse Detection
Xiaokun Luan, Xiyue Zhang, Jingyi Wang +1
Model reuse techniques can reduce the resource requirements for training high-performance deep neural networks (DNNs) by leveraging existing models. However, unauthorized reuse and…