{"id":"W4416037923","doi":"10.18653/v1/2025.codi-1.13","title":"Discourse Relation Recognition with Language Models Under Different Data Availability","year":2025,"lang":"","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Relation (database); Language model; Natural language; Feature (linguistics); Language identification; Data modeling","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004844777,0.0011006,0.001711677,0.001506104,0.001006673,0.003234731,0.00171404,0.001566591,0.003489398],"category_scores_gemma":[0.02156784,0.0006024244,0.001520093,0.001661757,0.0006227793,0.006900485,0.00192317,0.002691038,0.002316048],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001014446,"about_ca_system_score_gemma":0.002187956,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009848559,"about_ca_topic_score_gemma":0.008698337,"domain_scores_codex":[0.9959993,0.001394399,0.0003255295,0.001336481,0.0005138131,0.0004305023],"domain_scores_gemma":[0.9864743,0.009269358,0.0003273416,0.002390284,0.001213439,0.0003250802],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.006569726,0.001396402,0.01402386,0.00091225,0.0006914921,0.0007120143,0.001523691,0.173232,0.04707537,0.0108032,0.01619007,0.72687],"study_design_scores_gemma":[0.00006941765,0.00008621231,0.001508157,0.0000219512,0.0001416544,0.00007874185,0.0003240047,0.9726458,0.01318648,0.0101594,0.001733596,0.00004446671],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3779109,0.002252023,0.5942907,0.002665231,0.0005758154,0.0002470925,0.005452249,0.01243619,0.004169832],"genre_scores_gemma":[0.8894743,0.0004630685,0.1005979,0.000168306,0.000211601,0.0001916607,0.006262531,0.0004757456,0.002154827],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009848559,"threshold_uncertainty_score":0.02562189,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08687933868806245,"score_gpt":0.3117691931922972,"score_spread":0.2248898545042348,"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."}}