Shieldlm: Empowering Llms As Aligned, Customizable And Explainable Safety Detectors

Zhang Zhexin, Lu Yida, Ma Jingyuan, Zhang Di, Li Rui, Ke Pei, Sun Hao, Sha Lei, Sui Zhifang, Wang Hongning, Huang Minlie. Arxiv 2024

[Paper] [Code]    
Attention Mechanism Has Code Interpretability And Explainability Model Architecture Reinforcement Learning Responsible AI

The safety of Large Language Models (LLMs) has gained increasing attention in recent years, but there still lacks a comprehensive approach for detecting safety issues within LLMs’ responses in an aligned, customizable and explainable manner. In this paper, we propose ShieldLM, an LLM-based safety detector, which aligns with general human safety standards, supports customizable detection rules, and provides explanations for its decisions. To train ShieldLM, we compile a large bilingual dataset comprising 14,387 query-response pairs, annotating the safety of responses based on various safety standards. Through extensive experiments, we demonstrate that ShieldLM surpasses strong baselines across four test sets, showcasing remarkable customizability and explainability. Besides performing well on standard detection datasets, ShieldLM has also been shown to be effective in real-world situations as a safety evaluator for advanced LLMs. We release ShieldLM at \url{https://github.com/thu-coai/ShieldLM} to support accurate and explainable safety detection under various safety standards, contributing to the ongoing efforts to enhance the safety of LLMs.

Similar Work