ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark
This work addresses the problem of limited resources for Chinese harmful content detection, providing a benchmark and tools to improve efficiency and accuracy for content moderators, though it is incremental in building on existing detection methods.
The authors tackled the scarcity of Chinese datasets for harmful content detection by creating a comprehensive, professionally annotated benchmark from real-world data, which includes a knowledge rule base and a knowledge-augmented baseline that enables smaller models to achieve performance comparable to state-of-the-art LLMs.
Large language models (LLMs) have been increasingly applied to automated harmful content detection tasks, assisting moderators in identifying policy violations and improving the overall efficiency and accuracy of content review. However, existing resources for harmful content detection are predominantly focused on English, with Chinese datasets remaining scarce and often limited in scope. We present a comprehensive, professionally annotated benchmark for Chinese content harm detection, which covers six representative categories and is constructed entirely from real-world data. Our annotation process further yields a knowledge rule base that provides explicit expert knowledge to assist LLMs in Chinese harmful content detection. In addition, we propose a knowledge-augmented baseline that integrates both human-annotated knowledge rules and implicit knowledge from large language models, enabling smaller models to achieve performance comparable to state-of-the-art LLMs. Code and data are available at https://github.com/zjunlp/ChineseHarm-bench.