EMNLP 2024industry7 citations

SLM as Guardian: Pioneering AI Safety with Small Language Model

Ohjoon Kwon, Donghyeon Jeon, Nayoung Choi, Gyu-Hwung Cho, Hwiyeol Jo, Changbong Kim, Hyunwoo Lee, Inho Kang

Abstract

Most prior safety research of large language models (LLMs) has focused on enhancing the alignment of LLMs to better suit the safety requirements of their use cases. However, internalizing such safeguard features into larger models brought challenges of higher training cost and unintended degradation of helpfulness. 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.

BibTeX
@inproceedings{kwon-etal-2024-slm,
    title = "{SLM} as Guardian: Pioneering {AI} Safety with Small Language Model",
    author = "Kwon, Ohjoon  and
      Jeon, Donghyeon  and
      Choi, Nayoung  and
      Cho, Gyu-Hwung  and
      Jo, Hwiyeol  and
      Kim, Changbong  and
      Lee, Hyunwoo  and
      Kang, Inho  and
      Kim, Sun  and
      Park, Taiwoo",
    editor = "Dernoncourt, Franck  and
      Preo{\c{t}}iuc-Pietro, Daniel  and
      Shimorina, Anastasia",
    booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track",
    month = nov,
    year = "2024",
    address = "Miami, Florida, US",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.emnlp-industry.99/",
    doi = "10.18653/v1/2024.emnlp-industry.99",
    pages = "1333--1350"
}
SLM as Guardian: Pioneering AI Safety with Small Language Model · EMNLP 2024