{"id":"W4416437085","doi":"10.48550/arxiv.2511.02824","title":"Kosmos: An AI Scientist for Autonomous Discovery","year":2025,"lang":"","type":"preprint","venue":"ArXiv.org","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institutes of Health; Engineering and Physical Sciences Research Council; UK Dementia Research Institute; Medical Research Council; Cure Alzheimer's Fund; Canadian Institute for Advanced Research; Toyota Research Institute","keywords":"Limiting; Code (set theory); Reading (process); Scientific discovery; Source code","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.0007669389,0.0007945876,0.0008493085,0.0001626661,0.0005576788,0.0003246738,0.001569406,0.001741051,0.00006210095],"category_scores_gemma":[0.000770219,0.0007675113,0.0006726744,0.0002324587,0.001242925,0.00001818036,0.001873767,0.0007408705,0.0000339418],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000827948,"about_ca_system_score_gemma":0.001508524,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008075032,"about_ca_topic_score_gemma":0.0003316932,"domain_scores_codex":[0.9952071,0.0001860851,0.0008978097,0.002368369,0.0002829406,0.001057674],"domain_scores_gemma":[0.9969172,0.00009686831,0.0003937358,0.001922598,0.0003208191,0.0003487018],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.002325421,0.004299912,0.3718811,0.004627911,0.003385126,0.0001075433,0.001123942,0.0005703212,0.133238,0.001628138,0.0666645,0.4101481],"study_design_scores_gemma":[0.00184307,0.001648456,0.03856977,0.0005002925,0.0004177844,0.00001973191,0.0003420462,0.0006876704,0.05258016,0.0007392923,0.9010991,0.001552591],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9294108,0.004570935,0.05147139,0.00279025,0.007080229,0.001167675,0.001845213,0.0001160328,0.001547428],"genre_scores_gemma":[0.9557353,0.0009715816,0.004318365,0.002494903,0.001551777,0.0003630628,0.002947539,0.00006213142,0.03155536],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8344346,"threshold_uncertainty_score":0.9995549,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03738130673923349,"score_gpt":0.333183693005038,"score_spread":0.2958023862658045,"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."}}