用低秩适配器融合提升大模型安全性，有害回复减少42%。
Enhancing AI Safety Through the Fusion of Low Rank Adapters
- 将任务适配器与安全适配器通过低秩融合，动态调节输出
- 在基准测试中有害响应率降低42%，保留多指令处理能力
- 适合关注模型安全性的研究人员与工业部署团队
大语言模型的指令微调虽能提升特定任务表现，但可能引发对恶意提示生成有害回应的问题。本文探索使用低秩适配器融合（LoRA）来缓解此类风险，同时保持模型对多样化指令的处理能力。通过在知名基准数据集上与主流基线对比，我们证明：将任务适配器与专门在自建安全数据集上训练的安全适配器进行LoRA融合，可使有害响应率降低42%。然而，也观察到过度安全行为，即模型会拒绝与不安全提示相似的正常提示。
原文摘要 · Abstract (English)
Instruction fine-tuning of large language models (LLMs) is a powerful method for improving task-specific performance, but it can inadvertently lead to a phenomenon where models generate harmful responses when faced with malicious prompts. In this paper, we explore Low-Rank Adapter Fusion (LoRA) as a means to mitigate these risks while preserving the model's ability to handle diverse instructions effectively. Through an extensive comparative analysis against established baselines using recognized benchmark datasets, we demonstrate a 42\% reduction in the harmfulness rate by leveraging LoRA fusion between a task adapter and a safety adapter, the latter of which is specifically trained on our safety dataset. However, we also observe exaggerated safety behaviour, where the model rejects safe prompts that closely resemble unsafe ones
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。