用单个神经元实现低成本高效安全对齐，提升大模型输出安全性。
Light Alignment Improves LLM Safety via Model Self-Reflection with a Single Neuron
- 仅用一个神经元作门控，结合模型自省与外部引导。
- 训练开销低，跨模型规模泛化能力强，安全与性能兼顾。
- 适合追求轻量化部署的工业级大模型安全优化场景。
大语言模型（LLMs）的安全性已成为其发展中的关键问题。现有安全对齐方法多依赖后训练，计算成本高且泛化能力差。少数轻量级方法或依赖预先计算的安全注入，或过度依赖模型自身能力，导致泛化受限、生成效率下降。本文提出一种安全感知解码方法，仅需低成本训练专家模型，并使用单个神经元作为门控机制。通过有效平衡模型内在能力与外部引导，该方法在保持实用性的前提下显著提升输出安全性。实验表明，该方法在训练开销和跨模型规模泛化方面均具明显优势，为大模型的安全高效部署提供了新思路。代码已开源：https://github.com/Beijing-AISI/NGSD。
原文摘要 · Abstract (English)
The safety of large language models (LLMs) has increasingly emerged as a fundamental aspect of their development. Existing safety alignment for LLMs is predominantly achieved through post-training methods, which are computationally expensive and often fail to generalize well across different models. A small number of lightweight alignment approaches either rely heavily on prior-computed safety injections or depend excessively on the model's own capabilities, resulting in limited generalization and degraded efficiency and usability during generation. In this work, we propose a safety-aware decoding method that requires only low-cost training of an expert model and employs a single neuron as a gating mechanism. By effectively balancing the model's intrinsic capabilities with external guidance, our approach simultaneously preserves utility and enhances output safety. It demonstrates clear advantages in training overhead and generalization across model scales, offering a new perspective on lightweight alignment for the safe and practical deployment of large language models. Code: https://github.com/Beijing-AISI/NGSD.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。