arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.06887cs.CR
收藏

柏拉图的形:通过原型表示向 LLMs 提供后门防御即服务

Plato's Form: Toward Backdoor Defense-as-a-Service for LLMs with Prototype Representations

Nanyang Technological University(南洋理工大学);Wuhan University(武汉大学)

Chen Chen, Yuchen Sun, Jiaxin Gao, Yanwen Jia, Xueluan Gong, Qian Wang, Kwok-Yan Lam

AI总结:

PROTOPURIFY通过原型表示实现LLMs的后门防御即服务,有效降低后门攻击成功率并保持模型性能。

摘要:

大型语言模型(LLMs)正在越来越多地应用于安全敏感的应用,但仍然容易受到后门攻击的威胁。然而,现有的后门防御方法难以用于后门防御即服务(BDaaS),因为它们需要不切实际的侧信息(例如下游干净数据、已知触发器/目标或任务领域特定信息),并且缺乏可重用、可扩展的净化能力,以适应各种后门模型。在本文中,我们提出了PROTOPURIFY,一种通过参数编辑在最小假设下进行后门净化的框架。PROTOPURIFY首先从干净和被后门污染的模型对中构建后门向量池,将向量聚合为候选原型,并通过相似性匹配选择最符合目标模型的候选原型。PROTOPURIFY然后通过逐层原型对齐确定一个边界层,并通过抑制受影响层中的原型对齐组件进行针对性净化,从而实现细粒度的缓解,同时对良性效用的影响极小。设计为适用于BDaaS的原始构件,PROTOPURIFY支持重用性、可定制性、可解释性和运行时效率。在各种LLM上进行的实验表明,在分类和生成任务中,PROTOPURIFY在对抗6种不同攻击(包括单触发、多触发和无触发后门设置)时,始终优于6种代表性防御方法。PROTOPURIFY将ASR降低到低于10%,在某些情况下甚至低至1.6%,同时在干净效用上仅导致不到3%的下降。PROTOPURIFY进一步展示了对自适应后门变种的鲁棒性和在非后门模型上的稳定性。

英文摘要:

Large language models (LLMs) are increasingly deployed in security-sensitive applications, yet remain vulnerable to backdoor attacks. However, existing backdoor defenses are difficult to operationalize for Backdoor Defense-as-a-Service (BDaaS), as they require unrealistic side information (e.g., downstream clean data, known triggers/targets, or task domain specifics), and lack reusable, scalable purification across diverse backdoored models. In this paper, we present PROTOPURIFY, a backdoor purification framework via parameter edits under minimal assumptions. PROTOPURIFY first builds a backdoor vector pool from clean and backdoored model pairs, aggregates vectors into candidate prototypes, and selects the most aligned candidate for the target model via similarity matching. PROTOPURIFY then identifies a boundary layer through layer-wise prototype alignment and performs targeted purification by suppressing prototype-aligned components in the affected layers, achieving fine-grained mitigation with minimal impact on benign utility. Designed as a BDaaS-ready primitive, PROTOPURIFY supports reusability, customizability, interpretability, and runtime efficiency. Experiments across various LLMs on both classification and generation tasks show that PROTOPURIFY consistently outperforms 6 representative defenses against 6 diverse attacks, including single-trigger, multi-trigger, and triggerless backdoor settings. PROTOPURIFY reduces ASR to below 10%, and even as low as 1.6% in some cases, while incurring less than a 3% drop in clean utility. PROTOPURIFY further demonstrates robustness against adaptive backdoor variants and stability on non-backdoored models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑