Large Language Models (LLMs) have surged in popularity in recent months, but they have demonstrated concerning capabilities to generate harmful content when manipulated. While techniques like safety fine-tuning aim to minimize harmful use, recent works have shown that LLMs remain vulnerable to attacks that elicit toxic responses. In this work, we introduce the Proxy-Guided Attack on LLMs (PAL), the first optimization-based attack on LLMs in a black-box query-only setting. In particular, it relies on a surrogate model to guide the optimization and a sophisticated loss designed for real-world LLM APIs. Our attack achieves 84% attack success rate (ASR) on GPT-3.5-Turbo and 48% on Llama-2-7B, compared to 4% for the current state of the art. We also propose GCG++, an improvement to the GCG attack that reaches 94% ASR on white-box Llama-2-7B, and the Random-Search Attack on LLMs (RAL), a strong but simple baseline for query-based attacks. We believe the techniques proposed in this work will enable more comprehensive safety testing of LLMs and, in the long term, the development of better security guardrails. The code can be found at https://github.com/chawins/pal.
Ten Words Only Still Help: Improving Black-Box AI-Generated Text Detection via Proxy-Guided Efficient Re-Sampling
Key Lab of Intelligent Information Processing of Chinese Academy of Sciences（中国科学院智能信息处理重点实验室）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Xi’an Jiaotong University（西安交通大学）
Revitalizing Black-Box Interpretability: Actionable Interpretability for LLMs via Proxy Models
Key Lab of High Confidence Software Technologies (Peking University), Ministry of Education（高可信软件技术教育部重点实验室（北京大学））；School of Computer Science, Peking University（北京大学计算机学院）
Estimating the Black-box LLM Uncertainty with Distribution-Aligned Adversarial Distillation
School of Computer Science and Technology, Tianjin University, China（天津大学计算机科学与技术学院）；School of Future Technology, Tianjin University, China（天津大学未来技术学院）；Georgia Tech Shenzhen Institute, Tianjin University, China（Georgia Tech深圳研究院）；School of Artificial Intelligence, Tianjin University, China（天津大学人工智能学院）
Humanizing the Machine: Proxy Attacks to Mislead LLM Detectors
The Pennsylvania State University（宾夕法尼亚州立大学）；University of California, Los Angeles（加利福尼亚大学洛杉矶分校）；Nanjing University（南京大学）；University of Virginia（弗吉尼亚大学）；NEC Laboratories America（美国NEC实验室）
Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）；National Computer Network Emergency Response Technical Team, Coordination Center of China (CNCERT/CC)（国家计算机网络应急技术处理协调中心）
