{"id":"W7084109505","doi":"10.6084/m9.figshare.30204920","title":"Additional file 2 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":"article","venue":"Figshare","topic":"Vibrio bacteria research studies","field":"Biochemistry, Genetics and Molecular Biology","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":["metaresearch","research_integrity","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002152382,0.001893349,0.001101059,0.002960972,0.0007817179,0.001801865,0.002188371,0.001841984,0.7064706],"category_scores_gemma":[0.01863381,0.0008028651,0.001739337,0.002506233,0.0003102459,0.001973532,0.001695392,0.001296993,0.1409128],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001245277,"about_ca_system_score_gemma":0.001820941,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00739721,"about_ca_topic_score_gemma":0.01527221,"domain_scores_codex":[0.9992705,0.0001764965,0.0001323778,0.0002013037,0.0001472926,0.00007210933],"domain_scores_gemma":[0.9790269,0.01821634,0.0006013623,0.0007099361,0.001045732,0.000399701],"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":[0.0004381808,0.0001101546,0.003557795,0.003301668,0.0001078635,0.0002346172,0.0001316751,0.001847324,0.0004384584,0.002328863,0.9757642,0.01173915],"study_design_scores_gemma":[0.002216225,0.000191836,0.01270784,0.001567925,0.0003133427,0.0005978309,0.0003801119,0.01200031,0.002592457,0.02462433,0.9426165,0.0001913082],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.0002540565,0.00002327449,0.002062711,0.0001084717,0.00002765834,0.00005452089,0.9929823,0.003373316,0.001113717],"genre_scores_gemma":[0.007158163,0.00009421736,0.01682494,0.0004137815,0.00005402425,0.0008955567,0.966668,0.003438601,0.00445274],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.998158,"threshold_uncertainty_score":0.418684,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1727699109888409,"score_gpt":0.4119673493965754,"score_spread":0.2391974384077344,"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."}}