结构安全泛化问题
The Structural Safety Generalization Problem
AI总结:
针对LLM越狱安全挑战,提出研究安全机制在语义等价输入间无法泛化的失效机制,通过红队测试揭示多轮、多图像及翻译攻击漏洞,并提出结构重写护栏显著提升有害输入拒绝率且不过度拒绝良性输入。
摘要:
LLM越狱是一个普遍的安全挑战。鉴于该问题尚未变得易于处理,我们建议针对一个关键失效机制:安全机制无法在语义等价的输入间进行泛化。我们进一步通过要求所研究攻击具备理想的易处理属性来聚焦该目标:可解释性、模型间的可迁移性以及目标间的可迁移性。我们在此框架内进行红队测试,揭示了多轮、多图像和基于翻译的攻击的新漏洞。根据我们的设计,这些攻击在语义上与其单轮、单图像或未翻译的对应形式等价,从而实现系统比较;我们表明不同的结构会产生不同的安全结果。随后,我们通过提出一种结构重写护栏来展示该框架在促成新防御方面的潜力,该护栏将输入转换为更有利于安全评估的结构。该护栏显著提升了对有害输入的拒绝率,同时没有过度拒绝良性输入。因此,通过构建这一中间挑战——比通用防御更易处理,但对长期安全至关重要——我们突出了AI安全研究的一个关键里程碑。
英文摘要:
LLM jailbreaks are a widespread safety challenge. Given this problem has not yet been tractable, we suggest targeting a key failure mechanism: the failure of safety to generalize across semantically equivalent inputs. We further focus the target by requiring desirable tractability properties of attacks to study: explainability, transferability between models, and transferability between goals. We perform red-teaming within this framework by uncovering new vulnerabilities to multi-turn, multi-image, and translation-based attacks. These attacks are semantically equivalent by our design to their single-turn, single-image, or untranslated counterparts, enabling systematic comparisons; we show that the different structures yield different safety outcomes. We then demonstrate the potential for this framework to enable new defenses by proposing a Structure Rewriting Guardrail, which converts an input to a structure more conducive to safety assessment. This guardrail significantly improves refusal of harmful inputs, without over-refusing benign ones. Thus, by framing this intermediate challenge - more tractable than universal defenses but essential for long-term safety - we highlight a critical milestone for AI safety research.
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。