{"id":"W7131422250","doi":"10.1109/icdm65498.2025.00102","title":"Retrieval-Augmented Feature Generation for Domain-Specific Classification","year":2025,"lang":"","type":"article","venue":"","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"National Science Foundation","keywords":"Interpretability; Feature (linguistics); Domain (mathematical analysis); Feature vector; Domain knowledge; Feature model; Pattern recognition (psychology)","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":[0.00152303,0.001126377,0.00101699,0.002013865,0.000428963,0.0006604745,0.001804974,0.001017245,0.002688662],"category_scores_gemma":[0.005090476,0.0002988896,0.001314842,0.001561911,0.0006009524,0.001851211,0.001238109,0.001418425,0.001369801],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000642638,"about_ca_system_score_gemma":0.0007661943,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001828305,"about_ca_topic_score_gemma":0.003142422,"domain_scores_codex":[0.999135,0.0002481477,0.00005854406,0.0002927284,0.0001910126,0.00007457079],"domain_scores_gemma":[0.9976795,0.001164185,0.000150985,0.0006376454,0.0003115378,0.00005620982],"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":[0.0002727148,0.0003880912,0.002867401,0.0002859005,0.00010365,0.0002833988,0.0001583288,0.04684821,0.02155796,0.005189051,0.01363497,0.9084104],"study_design_scores_gemma":[0.00008754768,0.0002315688,0.001409739,0.00003406662,0.00008333433,0.0003473923,0.00008951559,0.9408949,0.02095984,0.02591646,0.009895994,0.00004963996],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02435478,0.000734778,0.9660214,0.000211562,0.00007174083,0.0001874931,0.0005054924,0.006858787,0.001053991],"genre_scores_gemma":[0.4040209,0.0003875543,0.5874462,0.0003169466,0.000107771,0.0004493535,0.004310252,0.000438426,0.002522725],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002688662,"threshold_uncertainty_score":0.00899452,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05892550197452328,"score_gpt":0.3007225011972353,"score_spread":0.241796999222712,"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."}}