FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Models

论文概览 论文标题:FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Models 数据规模:820条精心设计的指令,涵盖50+个NLP任务 核心创新:首创多层级细粒度约束遵循评估框架 ...

2025年03月27日 · 9 分钟 · 4458 字 · ZhaoYang