{"id":"W4412163903","doi":"10.1158/1557-3265.aimachine-b021","title":"Abstract B021: Current oncological large language model research lacks reproducibility, transparency, and long term support","year":2025,"lang":"en","type":"article","venue":"Clinical Cancer Research","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Term (time); Transparency (behavior); Current (fluid); Reproducibility; Medicine; Medical physics; Computer science; Intensive care medicine; Engineering; Computer security; Physics; Statistics; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1883137,0.00124382,0.00243641,0.007712926,0.002309103,0.01971371,0.004624129,0.003277137,0.04165471],"category_scores_gemma":[0.5771513,0.001594307,0.00455542,0.0145696,0.004380554,0.01991601,0.008130401,0.005420762,0.01806262],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004896058,"about_ca_system_score_gemma":0.02327705,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005210612,"about_ca_topic_score_gemma":0.00437216,"domain_scores_codex":[0.8549892,0.08822163,0.01861586,0.009773551,0.02650913,0.001890596],"domain_scores_gemma":[0.2441146,0.5709214,0.04151999,0.08164598,0.05803122,0.003766777],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.002095282,0.0002160312,0.02418131,0.05912901,0.002539902,0.0005785393,0.005427884,0.004174226,0.002772295,0.167839,0.2942282,0.4368184],"study_design_scores_gemma":[0.0005425683,0.0004040062,0.01353192,0.04321608,0.002015375,0.001051827,0.002972118,0.01198168,0.004385785,0.217634,0.7019506,0.0003142604],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04697392,0.1297209,0.3467606,0.2547934,0.0114524,0.003641638,0.08988104,0.01112324,0.1056528],"genre_scores_gemma":[0.425514,0.07165401,0.3060272,0.05832694,0.01055938,0.009990726,0.08828793,0.01160942,0.01803041],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8116863,"threshold_uncertainty_score":0.9959101,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3236301449901399,"score_gpt":0.6342184687243343,"score_spread":0.3105883237341944,"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."}}