通过随机平滑嵌入提升大模型抗攻击能力
Smoothed Embeddings for Robust Language Models
- 在生成每步词元时对嵌入向量加噪并聚合
- 相比基线防御，显著提升鲁棒性与可用性平衡
- 适合关注模型安全性的研究人员和开发者
提升大语言模型的安全性和可靠性是实现可信AI系统的关键。尽管对齐方法旨在抑制有害内容生成，但模型仍易受对抗性输入的越狱攻击，导致产生有害输出。本文提出随机嵌入平滑与词元聚合（RESTA）防御机制，在生成每个输出词元时对嵌入向量添加随机噪声并进行聚合，以更好地保留语义信息。实验表明，该方法在鲁棒性与可用性权衡上优于基线防御。
原文摘要 · Abstract (English)
Improving the safety and reliability of large language models (LLMs) is a crucial aspect of realizing trustworthy AI systems. Although alignment methods aim to suppress harmful content generation, LLMs are often still vulnerable to jailbreaking attacks that employ adversarial inputs that subvert alignment and induce harmful outputs. We propose the Randomized Embedding Smoothing and Token Aggregation (RESTA) defense, which adds random noise to the embedding vectors and performs aggregation during the generation of each output token, with the aim of better preserving semantic information. Our experiments demonstrate that our approach achieves superior robustness versus utility tradeoffs compared to the baseline defenses.
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。