CHiSafetyBench: A Chinese Hierarchical Safety Benchmark for Large Language Models

TL;DR

CHiSafetyBench is a comprehensive Chinese safety benchmark for large language models, assessing their ability to identify risky content and refuse risky questions across hierarchical safety categories, filling a crucial gap in Chinese safety evaluation.

Contribution

This work introduces the first Chinese safety benchmark with a hierarchical taxonomy, covering diverse safety categories and tasks, and validates automatic evaluation as effective for safety assessment.

Findings

Models show varied safety performance across domains.

Automatic evaluation correlates well with human judgment.

All models need improvement in Chinese safety capabilities.

Abstract

With the profound development of large language models(LLMs), their safety concerns have garnered increasing attention. However, there is a scarcity of Chinese safety benchmarks for LLMs, and the existing safety taxonomies are inadequate, lacking comprehensive safety detection capabilities in authentic Chinese scenarios. In this work, we introduce CHiSafetyBench, a dedicated safety benchmark for evaluating LLMs' capabilities in identifying risky content and refusing answering risky questions in Chinese contexts. CHiSafetyBench incorporates a dataset that covers a hierarchical Chinese safety taxonomy consisting of 5 risk areas and 31 categories. This dataset comprises two types of tasks: multiple-choice questions and question-answering, evaluating LLMs from the perspectives of risk content identification and the ability to refuse answering risky questions respectively. Utilizing this benchmark, we validate the feasibility of automatic evaluation as a substitute for human evaluation and conduct comprehensive automatic safety assessments on mainstream Chinese LLMs. Our experiments reveal the varying performance of different models across various safety domains, indicating that all models possess considerable potential for improvement in Chinese safety capabilities. Our dataset is publicly available at https://github.com/UnicomAI/UnicomBenchmark/tree/main/CHiSafetyBench.