SuperCLUE-Math6: Graded Multi-Step Math Reasoning Benchmark for LLMs in Chinese
This addresses the problem of limited evaluation tools for Chinese mathematical reasoning in AI, though it is incremental as an upgraded version of an existing dataset.
The authors tackled the lack of a comprehensive benchmark for evaluating mathematical reasoning in Chinese language models by introducing SuperCLUE-Math6, a dataset of over 2000 multi-step word problems, and found that top models like GPT-4 showed superior performance in experiments on 13 models.
We introduce SuperCLUE-Math6(SC-Math6), a new benchmark dataset to evaluate the mathematical reasoning abilities of Chinese language models. SC-Math6 is designed as an upgraded Chinese version of the GSM8K dataset with enhanced difficulty, diversity, and application scope. It consists of over 2000 mathematical word problems requiring multi-step reasoning and providing natural language solutions. We propose an innovative scheme to quantify the reasoning capability of large models based on performance over problems with different reasoning steps. Experiments on 13 representative Chinese models demonstrate a clear stratification of reasoning levels, with top models like GPT-4 showing superior performance. SC-Math6 fills the gap in Chinese mathematical reasoning benchmarks and provides a comprehensive testbed to advance the intelligence of Chinese language models.