PhonicsStoryBench
收藏官方服务：
资源简介：
PhonicsStoryBench is a large-scale benchmark dataset of LLM-generated phonics stories for early literacy education. The dataset is designed to evaluate the ability of large language models to generate text that satisfies multiple instructional constraints simultaneously. The dataset contains 39,096 short stories generated using six large language models across 16 prompt configurations.
提供机构：
Zenodo创建时间：
2026-03-30
