evaluation-task · Source-linked discovery

MMLU-Pro: Advanced Multitask Knowledge and Reasoning Evaluation

An advanced benchmark that tests both broad knowledge and reasoning capabilities across many subjects, featuring challenging questions and multiple-choice answers with increased difficulty and complexity.

Open interactive record →
OriginYubo Wang, Xueguang Ma, Ge Zhang et al.TopicsGeneral capabilityStatusimported

Can support

Not independently assessed by FronteraEval yet.

Cannot support by itself

No inference beyond the upstream source should be made until the protocol is reviewed.

Original sources