{"id":"W7084071816","doi":"10.6084/m9.figshare.30204917.v1","title":"Additional file 1 of A large language model-based tool for identifying relationships to industry in research on the carcinogenicity of benzene, cobalt, and aspartame","year":2025,"lang":"en","type":"dataset","venue":"Figshare","topic":"Healthcare Systems and Reforms","field":"Economics, Econometrics and Finance","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Occupational Cancer Research Centre; University of Toronto","funders":"","keywords":"Aspartame; Quality (philosophy); Identification (biology); Class (philosophy)","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001815919,0.001252205,0.001133652,0.003210895,0.0006333189,0.001634163,0.002069162,0.001933366,0.4445228],"category_scores_gemma":[0.01781824,0.0006504987,0.001426056,0.005376779,0.0002776048,0.001262524,0.001290484,0.001521061,0.0996125],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001757201,"about_ca_system_score_gemma":0.002414198,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02442081,"about_ca_topic_score_gemma":0.04961473,"domain_scores_codex":[0.9990076,0.0002469299,0.0001945408,0.0002699361,0.0001430488,0.0001379357],"domain_scores_gemma":[0.9869529,0.009471954,0.0009991177,0.0008587777,0.001276243,0.0004410484],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001394727,0.00005808679,0.003742751,0.00123482,0.00006620207,0.00003370582,0.00003061752,0.0007472089,0.00005576044,0.000862741,0.9905352,0.002493556],"study_design_scores_gemma":[0.002848187,0.0001511554,0.02923147,0.001528265,0.0003038301,0.0002122389,0.0002913344,0.003277812,0.0005399467,0.009148078,0.9523612,0.0001064984],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00007688662,0.0000128015,0.00005815806,0.00002952301,0.000004472155,0.00000934702,0.9995777,0.00005048838,0.0001806524],"genre_scores_gemma":[0.001546585,0.00004171888,0.0006394403,0.0001088358,0.00001354637,0.0002445346,0.9963169,0.00008056931,0.0010078],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.4445228,"threshold_uncertainty_score":0.7923207,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3290594644763958,"score_gpt":0.3942869679139518,"score_spread":0.06522750343755596,"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."}}