We introduce FollowBench, a Multi-level Fine-grained Constraints Following Benchmark for systemically and precisely evaluate the instruction-following capability of LLMs.
FollowBench comprehensively includes five different types (i.e., Content, Situation, Style, Format, and Example) of fine-grained constraints.
To enable a precise constraint following estimation on diverse… See the full description on the dataset page:
https://huggingface.co/datasets/YuxinJiang/FollowBench.
