Built independently by an author, for readers. Read the story and support ChapterPal

keyword

optimization-based jailbreaking

Optimization-based jailbreaking is an adversarial attack methodology used against artificial intelligence systems, particularly large language models, that employs automated mathematical optimization algorithms to generate inputs engineered to bypass safety guardrails and alignment filters. Unlike manual prompt engineering or social engineering attacks, this approach treats safety evasion as a formal optimization problem, systematically searching for token sequences or prompt suffixes that maximize the likelihood of the model producing restricted or harmful responses. The technique typically utilizes loss functions, model gradients, or discrete search heuristics to iteratively evaluate and update candidate tokens until the targeted model overrides its refusal behavior and fulfills the prohibited request.

1 item

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

Xiaojun Jia, Tianyu Pang, Chao Du, Yihao Huang, Jindong Gu, Yang Liu, Xiaochun Cao, Min Lin

OrganizationsNanyang Technological UniversitySea AI LabSun Yat-sen UniversityUniversity of Oxford

Why you should read this

Proposes I-GCG, an optimization-based attack method that substantially accelerates Greedy Coordinate Gradient convergence through adaptive multi-coordinate updates and diverse target templates, achieving near-100% jailbreak success rates across language model benchmarks.

Large language models (LLMs) are being rapidly developed, and a key component of their widespread deployment is their safety-related alignment. Many red-teaming efforts aim to jailbreak LLMs, where among these efforts, the Greedy Coordinate Gradient (GCG) attack's success has led to a growing interest in the study of optimization-based jailbreaking techniques. Although GCG is a significant milestone, its attacking efficiency remains unsatisfactory. In this paper, we present several improved (empirical) techniques for optimization-based jailbreaks like GCG. We first observe that the single target template of "Sure" largely limits the attacking performance of GCG; given this, we propose to apply diverse target templates containing harmful self-suggestion and/or guidance to mislead LLMs. Besides, from the optimization aspects, we propose an automatic multi-coordinate updating strategy in GCG (i.e., adaptively deciding how many tokens to replace in each step) to accelerate convergence, as well as tricks like easy-to-hard initialisation. Then, we combine these improved technologies to develop an efficient jailbreak method, dubbed I-GCG. In our experiments, we evaluate on a series of benchmarks (such as NeurIPS 2023 Red Teaming Track). The results demonstrate that our improved techniques can help GCG outperform state-of-the-art jailbreaking attacks and achieve nearly 100% attack success rate. The code is released at this https URL.

Added

2026-09-26