{"kind":"task","effective_mode":"full","benchmark":{"kind":"benchmark","effective_mode":"full","slug":"mmlu-pro","formal_name":"MMLU-Pro","introduction":"MMLU-Pro rebuilds MMLU with ten answer options instead of four and removes items that no longer separate models. Questions span fourteen academic and professional subjects.","introduction_ja":"","introduction_en":"","category":"Category not supplied","task_count":null,"acquisition_status":"Acquisition status not supplied","official_url":"https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro","indexing_mode":"index","profile":{"resources":[],"task_format":"","scoring":"","metric":"","size":"","answer_access":"","license":"","citation":"","maintainer":"","released":"","why_hard":"","related":[],"history":"","construction":"","limitations":"","saturation":"","scores":[]}},"task_id":"7e638cad-deac-539b-8c7c-7396347dafce","task_key":"default--test--8247","task_revision_id":"1","upstream_id":"8247","short_description":"Marguerite earned a score between 75 and 89 on all of her previous spelling…","config":"default","split":"test","body":"{\"category\":\"math\",\"options\":\"['The median will decrease.', 'The mode will decrease.', 'The median will remain the same.', 'The mean will increase.', 'The range will decrease.', 'The mean will decrease.', 'The mean will remain the same.', 'The mode will increase.']\",\"question\":\"Marguerite earned a score between 75 and 89 on all of her previous spelling tests. She earned a score of 100 on her next test. Which of the following statements is true?\",\"src\":\"ori_mmlu-elementary_mathematics\"}","display_format":"text","language":"","answer_status":"published","assets":[],"source_url":"https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro","history":"initial import","indexing_mode":"index","subproblems":[],"grids":[]}