OASIS：针对硬标签黑盒文本攻击优化攻击者序列
OASIS: Optimizing Attacker Sequences for Hard-Label Black-Box Text Attacks
浏览论文内容
中文总结 AI 辅助
OASIS通过双目标攻击链搜索优化硬标签黑盒文本攻击的攻击者序列，在多数据集、多模型上性能优于基线，为提升攻击效果提供新优化方向。
中文摘要 AI 辅助
不同攻击方法遵循不同的搜索轨迹，在不同的样本子集上取得成功，而现有的硬标签黑盒文本攻击主要专注于改进单个攻击者或手动组合它们。我们提出了OASIS（Our Method），一种用于优化硬标签黑盒文本攻击中攻击者序列的方法。OASIS首先在候选序列上执行一次性双目标攻击链搜索，以平衡攻击成功率和扰动，随后在攻击链执行阶段复用所选的固定全局链。在多个数据集、受害模型和大语言模型上的实验表明，OASIS始终优于强大的独立基线和简单的手动构建链。这些结果表明，攻击者组合不仅仅是一种实现选择，而是提升硬标签黑盒文本攻击性能的可行优化目标。
英文摘要
Different attack methods follow different search trajectories, they succeed on different subsets of samples, whereas existing hard-label black-box text attacks mainly focus on improving individual attackers or manually combining them. We present OASIS, a method for optimizing attacker sequences in hard-label black-box text attacks. OASIS first performs a one-time bi-objective attack chain search over candidate sequences to balance attack success rate and perturbation, and then reuses the selected fixed global chain during attack chain execution. Experiments across multiple datasets, victim models, and large language models show that OASIS consistently outperforms strong standalone baselines and simple manually constructed chains. These results suggest that attacker composition is not merely an implementation choice, but a practical optimization target for improving hard-label black-box text attacks.
发表机构
- School of Information Science and Technology（信息科学与技术学院）
- Guangdong University of Foreign Studies（广东外语外贸大学）
机构由 AI 辅助整理，请以论文原文为准。
