{"id":"W7125631796","doi":"10.1109/cascon66301.2025.00074","title":"Revisiting Thread Disentanglement in Technical Chat with Fine-Tuned LLMs and Graph-Based Clustering","year":2025,"lang":"","type":"article","venue":"","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Royal Military College of Canada; Queen's University","funders":"","keywords":"Cluster analysis; Thread (computing); Beijing","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001361862,0.0003641009,0.0004550949,0.0004175222,0.0003278384,0.0004049128,0.0005088706,0.0001613864,0.0000744016],"category_scores_gemma":[0.00007262068,0.0003093434,0.0000865958,0.001780436,0.0001763737,0.000330848,0.0006412644,0.0004561677,0.000006022374],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001477532,"about_ca_system_score_gemma":0.0002007099,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008090249,"about_ca_topic_score_gemma":0.0004586638,"domain_scores_codex":[0.9970788,0.0002227829,0.0007331652,0.0009339484,0.0003700761,0.000661246],"domain_scores_gemma":[0.9987069,0.0002706761,0.0001711137,0.0005904434,0.00009012407,0.000170678],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007427124,0.0005003008,0.2935473,0.00188125,0.0001252005,0.0002736016,0.001267907,0.002945616,0.003557498,0.3152412,0.0001752183,0.3797422],"study_design_scores_gemma":[0.007018484,0.0006281328,0.05803046,0.008770437,0.0001289695,0.00004166136,0.0007373945,0.8970137,0.0195973,0.005116725,0.001209147,0.001707544],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02455908,0.001148253,0.957518,0.01334836,0.0002604297,0.000724789,0.000004880273,0.0001671425,0.002269059],"genre_scores_gemma":[0.9652883,0.00006395962,0.03314032,0.001087415,0.00004831105,0.0000415303,0.000007967105,0.00001195032,0.0003102935],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9407292,"threshold_uncertainty_score":0.9999359,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02370573416189972,"score_gpt":0.2934075419077976,"score_spread":0.2697018077458979,"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."}}