{"kind":"task","effective_mode":"full","benchmark":{"kind":"benchmark","effective_mode":"full","slug":"mmlu","formal_name":"MMLU","introduction":"MMLU asks four-way multiple-choice questions across 57 subjects, from elementary material to professional-level examinations. Since 2020 it has been the most widely cited single reference point for general capability.","introduction_ja":"","introduction_en":"","category":"Category not supplied","task_count":null,"acquisition_status":"Acquisition status not supplied","official_url":"https://github.com/hendrycks/test","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":"ce50f66f-fc4d-5ee1-b065-b06365abd1f9","task_key":"all--test--ce50f66f-fc4d-5ee1-b065-b06365abd1f9","task_revision_id":"1","upstream_id":"","short_description":"For a Gaussian Bayes classifier, which one of these structural assumptions is…","config":"all","split":"test","body":"{\"choices\":\"['Whether we learn the class centers by Maximum Likelihood or Gradient Descent', 'Whether we assume full class covariance matrices or diagonal class covariance matrices', 'Whether we have equal class priors or priors estimated from the data', 'Whether we allow classes to have different mean vectors or we force them to share the same mean vector']\",\"question\":\"For a Gaussian Bayes classifier, which one of these structural assumptions is the one that most affects the trade-off between underfitting and overfitting:\",\"subject\":\"machine_learning\"}","display_format":"text","language":"","answer_status":"published","assets":[],"source_url":"https://github.com/hendrycks/test","history":"initial import","indexing_mode":"index","subproblems":[],"grids":[]}