{"id":"W4416230956","doi":"10.26434/chemrxiv-2025-6dj8j","title":"Bringing ML to the real world: rewards are all we need","year":2025,"lang":"","type":"article","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Operationalization; Extant taxon; Function (biology); Key (lock); Measure (data warehouse); Human intelligence","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.004685986,0.0006441343,0.0007759748,0.0005178765,0.0009974441,0.00165673,0.003333707,0.0001845143,0.004079209],"category_scores_gemma":[0.001655347,0.0005133125,0.000203785,0.002748188,0.0005036751,0.0003535013,0.001901138,0.000578836,0.002343134],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003690761,"about_ca_system_score_gemma":0.0003827592,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000990541,"about_ca_topic_score_gemma":0.0004111404,"domain_scores_codex":[0.9943034,0.0005636523,0.001046288,0.001587322,0.0009476427,0.001551673],"domain_scores_gemma":[0.9962727,0.0004773113,0.0005354334,0.002111908,0.000235791,0.0003668841],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001912517,0.0001395058,0.0081852,0.0006130004,0.00005588572,0.0000530935,0.00383455,0.0086094,0.9163305,0.004502612,0.05483641,0.002648624],"study_design_scores_gemma":[0.0008119622,0.0001070248,0.03812614,0.002367172,0.0001996964,0.00001256926,0.001143736,0.01424181,0.4611007,0.001594679,0.4789397,0.001354831],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8451608,0.0005948417,0.002392143,0.09094436,0.01152051,0.001212047,0.00001438984,0.0004888489,0.047672],"genre_scores_gemma":[0.9433091,0.0002151003,0.003920646,0.004347116,0.0009191133,0.00009764025,0.000003452602,0.00006316304,0.04712471],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4552298,"threshold_uncertainty_score":0.9997318,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01820846254988943,"score_gpt":0.3046869061053663,"score_spread":0.2864784435554769,"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."}}