{"id":"W7091625457","doi":"10.5281/zenodo.17353962","title":"Data for \"A flaw in using pre-trained pLMs in protein-protein interaction inference models\"","year":2025,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Prenatal Screening and Diagnostics","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Inference; Pattern recognition (psychology); Feature (linguistics); Bayesian inference; Data modeling","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":[],"consensus_categories":[],"category_scores_codex":[0.004037805,0.002289857,0.001216389,0.001183382,0.001254361,0.001915749,0.003898465,0.004089079,0.05310791],"category_scores_gemma":[0.01564338,0.0007601068,0.002451732,0.001849369,0.001197966,0.001121783,0.002107495,0.003223696,0.05957792],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002051936,"about_ca_system_score_gemma":0.002951615,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02354408,"about_ca_topic_score_gemma":0.06545812,"domain_scores_codex":[0.9972425,0.0008064512,0.0002448545,0.0006403964,0.0008261206,0.0002396564],"domain_scores_gemma":[0.9927804,0.002644733,0.0004556231,0.001995793,0.001715797,0.0004075899],"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":[0.0001025021,0.0000377817,0.0003739873,0.000242772,0.00003582343,0.00001935301,0.000009392677,0.0004949912,0.00009801705,0.0004096255,0.9962083,0.001967472],"study_design_scores_gemma":[0.001138279,0.00005814703,0.004580825,0.0003245176,0.0000855555,0.0001645046,0.00006194395,0.003898024,0.00173454,0.005553558,0.9823376,0.00006248959],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0008475768,0.0002053332,0.0009788286,0.001404788,0.0004295851,0.00006411994,0.9922639,0.001591543,0.002214326],"genre_scores_gemma":[0.002382803,0.00005131839,0.002474303,0.0008206375,0.00003497201,0.0002259261,0.9914793,0.0002433238,0.002287403],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.05310791,"threshold_uncertainty_score":0.1776636,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1096985999752618,"score_gpt":0.3490661777948798,"score_spread":0.239367577819618,"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."}}