{"id":"W4409093390","doi":"10.2196/72998","title":"Citation Accuracy Challenges Posed by Large Language Models","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"scientometrics and bibliometrics research","field":"Decision Sciences","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"","keywords":"Citation; Computer science; Natural language processing; Library science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","bibliometrics","insufficient_payload"],"consensus_categories":["bibliometrics"],"category_scores_codex":[0.01029172,0.0001009563,0.0001771252,0.01615532,0.000153858,0.0005943072,0.001359817,0.0001928408,0.001563502],"category_scores_gemma":[0.06933685,0.00007334239,0.0000730652,0.04896206,0.00006758771,0.000612413,0.0002288589,0.000271255,0.0002849861],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001090099,"about_ca_system_score_gemma":0.001462048,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005402448,"about_ca_topic_score_gemma":0.00002099021,"domain_scores_codex":[0.9919006,0.0002439303,0.0004888201,0.0005145372,0.006485151,0.0003669444],"domain_scores_gemma":[0.994554,0.003012058,0.0001412355,0.0005327978,0.001342527,0.0004173156],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000008342809,0.000603653,0.0005268469,0.00001238575,0.000005900094,8.560449e-7,0.001312165,6.15614e-7,0.000214294,0.01572169,0.3137069,0.6678863],"study_design_scores_gemma":[0.00206771,0.0001581884,0.07155684,0.0001614919,0.00001422725,0.000005009521,0.02839205,0.08612143,0.0007013284,0.1683442,0.6419815,0.0004960135],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7020292,0.02720254,0.02608098,0.1140949,0.003556882,0.001120283,0.00004638881,0.0001341535,0.1257346],"genre_scores_gemma":[0.9878913,0.001033806,0.0002793981,0.002105538,0.0001299124,0.0001018946,0.00004392119,0.000006094242,0.008408145],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6673903,"threshold_uncertainty_score":0.9993492,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3294677681499928,"score_gpt":0.60707918867813,"score_spread":0.2776114205281372,"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."}}