{"id":"W7118165281","doi":"10.1109/aiccsa66935.2025.11315432","title":"A Novel Competency Tagging Method Through Semantic Search Using Fine-Tuned LLM","year":2025,"lang":"","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; Université TÉLUQ","funders":"","keywords":"Matching (statistics); Semantic matching; Embedding; Similarity (geometry); Adaptation (eye); Semantic similarity; Training set; Process (computing)","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":[0.00130664,0.001503747,0.001330227,0.004136945,0.0008246149,0.00139012,0.001519827,0.00139273,0.003463731],"category_scores_gemma":[0.005384458,0.0004060318,0.001535567,0.002497459,0.0006841375,0.003412292,0.002046987,0.001341752,0.004568615],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008726415,"about_ca_system_score_gemma":0.002334872,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005994636,"about_ca_topic_score_gemma":0.01091684,"domain_scores_codex":[0.9984871,0.0004444655,0.0001569422,0.0003864232,0.0003758398,0.0001491108],"domain_scores_gemma":[0.9985856,0.0004967053,0.0001375035,0.0002971092,0.0003898636,0.00009328375],"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":[0.0004556276,0.0006557776,0.00594813,0.0006265463,0.0001987876,0.0004828035,0.0004891217,0.04583422,0.05574306,0.008984269,0.02756393,0.8530178],"study_design_scores_gemma":[0.0001123041,0.00024851,0.002039041,0.00005832877,0.0001127579,0.0005572668,0.0003848581,0.941294,0.02791889,0.01514316,0.01203669,0.00009432402],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03995626,0.001100016,0.9439564,0.0004090778,0.0001681555,0.0002381262,0.001138936,0.01014537,0.002887791],"genre_scores_gemma":[0.3691113,0.0005492616,0.613419,0.0007934858,0.0002153493,0.0003927538,0.007150218,0.0009837538,0.007384912],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005994636,"threshold_uncertainty_score":0.0119195,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09893633220898373,"score_gpt":0.3777495625801211,"score_spread":0.2788132303711374,"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."}}