Tian Dong (Shanghai Jiao Tong University), Minhui Xue (CSIRO's Data61), Guoxing Chen (Shanghai Jiao Tong University), Rayne Holland (CSIRO's Data61), Yan Meng (Shanghai Jiao Tong University), Shaofeng Li (Southeast University), Zhen Liu (Shanghai Jiao Tong University), Haojin Zhu (Shanghai Jiao Tong University)

Open-source Large Language Models (LLMs) have recently gained popularity because of their comparable performance to proprietary LLMs. To efficiently fulfill domain-specialized tasks, open-source LLMs can be refined, without expensive accelerators, using low-rank adapters. However, it is still unknown whether low-rank adapters can be exploited to control LLMs. To address this gap, we demonstrate that an infected adapter can induce, on specific triggers, an LLM to output content defined by an adversary and to even maliciously use tools. To train a Trojan adapter, we propose two novel attacks, POLISHED and FUSION, that improve over prior approaches. POLISHED uses a superior LLM to align naïvely poisoned data based on our insight that it can better inject poisoning knowledge during training. In contrast, FUSION leverages a novel over-poisoning procedure to transform a benign adapter into a malicious one by magnifying the attention between trigger and target in model weights. In our experiments, we first conduct two case studies to demonstrate that a compromised LLM agent can use malware to control the system (e.g., a LLM-driven robot) or to launch a spear-phishing attack. Then, in terms of targeted misinformation, we show that our attacks provide higher attack effectiveness than the existing baseline and, for the purpose of attracting downloads, preserve or improve the adapter’s utility. Finally, we designed and evaluated three potential defenses. However, none proved entirely effective in safeguarding against our attacks, highlighting the need for more robust defenses supporting a secure LLM supply chain.

View More Papers

YuraScanner: Leveraging LLMs for Task-driven Web App Scanning

Aleksei Stafeev (CISPA Helmholtz Center for Information Security), Tim Recktenwald (CISPA Helmholtz Center for Information Security), Gianluca De Stefano (CISPA Helmholtz Center for Information Security), Soheil Khodayari (CISPA Helmholtz Center for Information Security), Giancarlo Pellegrino (CISPA Helmholtz Center for Information Security)

Read More

VoiceRadar: Voice Deepfake Detection using Micro-Frequency and Compositional Analysis

Kavita Kumari (Technical University of Darmstadt), Maryam Abbasihafshejani (University of Texas at San Antonio), Alessandro Pegoraro (Technical University of Darmstadt), Phillip Rieger (Technical University of Darmstadt), Kamyar Arshi (Technical University of Darmstadt), Murtuza Jadliwala (University of Texas at San Antonio), Ahmad-Reza Sadeghi (Technical University of Darmstadt)

Read More

Probe-Me-Not: Protecting Pre-trained Encoders from Malicious Probing

Ruyi Ding (Northeastern University), Tong Zhou (Northeastern University), Lili Su (Northeastern University), Aidong Adam Ding (Northeastern University), Xiaolin Xu (Northeastern University), Yunsi Fei (Northeastern University)

Read More

ProvGuard: Detecting SDN Control Policy Manipulation via Contextual Semantics...

Ziwen Liu (Beihang University), Jian Mao (Beihang University; Tianmushan Laboratory; Hangzhou Innovation Institute, Beihang University), Jun Zeng (National University of Singapore), Jiawei Li (Beihang University; National University of Singapore), Qixiao Lin (Beihang University), Jiahao Liu (National University of Singapore), Jianwei Zhuge (Tsinghua University; Zhongguancun Laboratory), Zhenkai Liang (National University of Singapore)

Read More