XIFBench: Evaluating Large Language Models on Multilingual Instruction Following
收藏Harvard Dataverse2025-05-14 更新2026-04-09 收录
下载链接：
https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/9EBYA4
下载链接
链接失效反馈官方服务：
资源简介：
We introduce XIFBench, a constraint-based benchmark for evaluating multilingual instruction-following in LLMs across languages with diverse resource levels.
创建时间：
2025-01-01
