{"id":"W2953087206","doi":"10.48550/arxiv.1906.07282","title":"The Cells Out of Sample (COOS) dataset and benchmarks for measuring out-of-sample generalization of image classifiers","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Cell Image Analysis Techniques","field":"Biochemistry, Genetics and Molecular Biology","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Sample (material); Generalization; Artificial intelligence; Pattern recognition (psychology); Computer science; Image (mathematics); Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.007243903,0.002342739,0.001098141,0.002799931,0.001147768,0.001677729,0.002360336,0.002516322,0.002102469],"category_scores_gemma":[0.02562251,0.0004083111,0.001638832,0.002585215,0.001685226,0.001984132,0.002066107,0.002687733,0.001580373],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001925801,"about_ca_system_score_gemma":0.001654761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008280916,"about_ca_topic_score_gemma":0.01301072,"domain_scores_codex":[0.9951885,0.000930517,0.0004723661,0.001413281,0.001675974,0.0003192996],"domain_scores_gemma":[0.9865261,0.004891983,0.001161692,0.004605692,0.002189974,0.0006245304],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003306003,0.002857442,0.1083244,0.003740533,0.002528531,0.00122142,0.0008590667,0.1835135,0.07315771,0.01284651,0.3072675,0.3003774],"study_design_scores_gemma":[0.0006441654,0.002599201,0.1122549,0.0004180932,0.0005092678,0.003428058,0.0007643425,0.619412,0.12272,0.03093903,0.1059817,0.0003292894],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"dataset","genre_scores_codex":[0.6605529,0.006867948,0.1764912,0.003273977,0.001699541,0.001742598,0.1054003,0.02514316,0.01882833],"genre_scores_gemma":[0.5869671,0.001367342,0.1487489,0.001436408,0.0004322188,0.001657771,0.250032,0.003056538,0.006301736],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.9927561,"threshold_uncertainty_score":0.03830987,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05735388509598731,"score_gpt":0.2279074514219108,"score_spread":0.1705535663259234,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}