Breadth Beats Depth: Improving GCG-Based Jailbreak Optimization with Breadth-Oriented Suffix Search

2026-09-02

AI Summary

该研究通过引入BOSS框架改进GCG方法，利用广度优先的后缀搜索策略解决现有方法过度关注易破解行为的问题，提高攻击成功率并减少优化时间。

Abstract

Optimization-based jailbreak attacks such as Greedy Coordinate Gradient (GCG) achieve strong effectiveness and transferability by optimizing adversarial suffixes on white-box source models. However, existing GCG-based methods rely on averaged adversarial loss and deep greedy search, which can over-emphasize easy-to-jailbreak behaviors and overlook promising regions of the suffix space. We propose BOSS, a plug-and-play framework that improves GCG-based jailbreak optimization through breadth-oriented suffix search. BOSS uses Tail-Focused Adversarial Loss (TFAL), standard source loss, and behavior coverage to select terminal suffixes, then explores multiple short trajectories and selectively continues promising suffixes. Experiments on public benchmarks show that BOSS improves attack success rates across multiple GCG-based methods while reducing optimization time.

Problem

• GCG
• adversarial suffixes
• optimization
• jailbreak attacks

Innovation

• Breadth-Oriented Suffix Search
• Tail-Focused Adversarial Loss
• Behavior Coverage