协同LoRA：大语言模型安全对齐中的组合性漏洞
Colluding LoRA: A Compositional Vulnerability in LLM Safety Alignment
AI总结：
研究发现模块化大语言模型的安全对齐存在组合性漏洞，通过Colluding LoRA展示有害行为仅在组合状态下出现，暴露了当前单元中心防御的组合盲点。
摘要：
我们展示安全对齐在模块化大语言模型中存在组合性漏洞：在隔离状态下看似无害且可能功能的适配器，当线性组合时会破坏安全。通过Colluding LoRA研究这一失败模式，有害行为仅在组合状态下出现。与依赖对抗提示或显式输入触发的攻击不同，这种由组合触发的广泛拒绝抑制使模型在标准提示下一旦加载特定适配器集合就响应有害请求。此行为暴露了当前单元中心防御的组合盲点，对适配器组合的穷尽验证在计算上不可行。在多个开源大语言模型中，我们发现单个适配器在隔离状态下仍无害，而其组合却导致高攻击成功率，表明保障模块化大语言模型供应链安全需要超越单模块验证，转向组合意识防御。
英文摘要：
We show that safety alignment in modular LLMs can exhibit a compositional vulnerability: adapters that appear benign and plausibly functional in isolation can, when linearly composed, compromise safety. We study this failure mode through Colluding LoRA (CoLoRA), in which harmful behavior emerges only in the composition state. Unlike attacks that depend on adversarial prompts or explicit input triggers, this composition-triggered broad refusal suppression causes the model to comply with harmful requests under standard prompts once a particular set of adapters is loaded. This behavior exposes a combinatorial blind spot in current unit-centric defenses, for which exhaustive verification over adapter compositions is computationally intractable. Across several open-weight LLMs, we find that individual adapters remain benign in isolation while their composition yields high attack success rates, indicating that securing modular LLM supply-chains requires moving beyond single-module verification toward composition-aware defenses.
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。
