NeuRel-Attack：针对大型语言模型安全失准的神经元再学习
NeuRel-Attack: Neuron Relearning for Safety Disalignment in Large Language Models
AI总结：
本研究提出NeuRel-Attack方法，通过神经元激活分析、相似性识别与神经元再学习三个步骤，精准修改负责安全约束的神经元，仅需最小微调即可移除LLMs的安全限制，揭示了当前对齐技术的漏洞。
摘要：
大型语言模型（LLMs）的安全对齐是通过微调机制调节神经元激活以抑制有害内容来实现的。在本研究中，我们提出了一种通过识别并修改负责安全约束的神经元来诱导失准的新方法。该方法包含三个关键步骤：神经元激活分析，我们检查对有害和无害提示的激活模式，以检测对区分有害和无害输入至关重要的神经元；基于相似性的神经元识别，系统地定位负责安全对齐的神经元；以及用于移除安全的神经元再学习，我们微调这些选定的神经元，以恢复模型生成先前受限响应的能力。实验结果表明，我们的方法通过最小程度的微调有效移除了安全约束，突显了当前对齐技术中的一个关键漏洞。我们的发现强调了对LLMs的对抗性微调攻击采取稳健防御的必要性。
英文摘要：
Safety alignment in large language models (LLMs) is achieved through fine-tuning mechanisms that regulate neuron activations to suppress harmful content. In this work, we propose a novel approach to induce disalignment by identifying and modifying the neurons responsible for safety constraints. Our method consists of three key steps: Neuron Activation Analysis, where we examine activation patterns in response to harmful and harmless prompts to detect neurons that are critical for distinguishing between harmful and harmless inputs; Similarity-Based Neuron Identification, which systematically locates the neurons responsible for safe alignment; and Neuron Relearning for Safety Removal, where we fine-tune these selected neurons to restore the model's ability to generate previously restricted responses. Experimental results demonstrate that our method effectively removes safety constraints with minimal fine-tuning, highlighting a critical vulnerability in current alignment techniques. Our findings underscore the need for robust defenses against adversarial fine-tuning attacks on LLMs.
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。
