arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.19795cs.CLcs.AI
收藏

SLM作为守护者:以小型语言模型引领AI安全

SLM as Guardian: Pioneering AI Safety with Small Language Models

Naver(纳弗(韩国互联网公司))

Ohjoon Kwon, Donghyeon Jeon, Nayoung Choi, Gyu-Hwung Cho, Changbong Kim, Hyunwoo Lee, Inho Kang, Sun Kim, Taiwoo Park

AI总结:

本文提出利用小型语言模型(SLM)作为模块化守护者,通过多任务学习同时进行有害查询检测和安全响应生成,以较低成本实现与大型模型相当或更优的安全性能。

摘要:

以往关于大型语言模型(LLM)的安全研究大多集中于增强LLM的对齐,以更好地满足人类的安全要求。然而,将此类安全特性内化到更大的模型中,带来了训练成本更高和有用性意外下降的挑战。为了克服这些挑战,采用模块化方法,利用较小的LLM来检测有害用户查询,被视为在设计具有安全要求的基于LLM的系统时的一种便捷解决方案。在本文中,我们利用较小的LLM同时进行有害查询检测和安全响应生成。我们介绍了我们的安全要求和有害性分类体系,然后提出了一种多任务学习机制,将两个任务融合到单个模型中。我们证明了我们方法的有效性,与公开可用的LLM相比,在有害查询检测和安全响应生成性能上达到了同等或超越的水平。

英文摘要:

Most prior safety research of large language models (LLMs) has focused on enhancing the alignment of LLMs to better suit the safety requirements of humans. However, internalizing such safeguard features into larger models brought challenges of higher training cost and unintended degradation of helpfulness. To overcome such challenges, a modular approach employing a smaller LLM to detect harmful user queries is regarded as a convenient solution in designing LLM-based system with safety requirements. In this paper, we leverage a smaller LLM for both harmful query detection and safeguard response generation. We introduce our safety requirements and the taxonomy of harmfulness categories, and then propose a multi-task learning mechanism fusing the two tasks into a single model. We demonstrate the effectiveness of our approach, providing on par or surpassing harmful query detection and safeguard response performance compared to the publicly available 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。

↑