Chinese SafetyQA: A Safety Short-form Factuality Benchmark for Large Language Models
This addresses safety concerns for LLM deployment in Chinese-speaking regions by providing a benchmark, though it is incremental as it adapts existing evaluation approaches to a specific domain.
The authors tackled the problem of evaluating safety knowledge factuality in large language models by introducing the Chinese SafetyQA benchmark, which assesses models' ability to answer short safety-related questions accurately in Chinese, and they performed comprehensive evaluations showing how this capability relates to other LLM abilities like RAG and robustness.
With the rapid advancement of Large Language Models (LLMs), significant safety concerns have emerged. Fundamentally, the safety of large language models is closely linked to the accuracy, comprehensiveness, and clarity of their understanding of safety knowledge, particularly in domains such as law, policy and ethics. This factuality ability is crucial in determining whether these models can be deployed and applied safely and compliantly within specific regions. To address these challenges and better evaluate the factuality ability of LLMs to answer short questions, we introduce the Chinese SafetyQA benchmark. Chinese SafetyQA has several properties (i.e., Chinese, Diverse, High-quality, Static, Easy-to-evaluate, Safety-related, Harmless). Based on Chinese SafetyQA, we perform a comprehensive evaluation on the factuality abilities of existing LLMs and analyze how these capabilities relate to LLM abilities, e.g., RAG ability and robustness against attacks.