{"id":"W7130700679","doi":"10.1109/swc65939.2025.00031","title":"Benchmarking Few-Shot Learning Techniques for Steel Surface Defect Detection","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Moncton","funders":"","keywords":"Benchmarking; Focus (optics); Categorization; Implementation; Transfer of learning; Quality (philosophy)","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","sts"],"consensus_categories":[],"category_scores_codex":[0.0006817369,0.000403965,0.0003746925,0.0002059329,0.001422113,0.0003905081,0.0009340636,0.0002360651,0.00002999888],"category_scores_gemma":[0.0001437016,0.0004386908,0.000325725,0.001977719,0.0001108379,0.0006863188,0.000597088,0.0005884083,0.00001877314],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002664283,"about_ca_system_score_gemma":0.0001330957,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002775386,"about_ca_topic_score_gemma":0.00009347124,"domain_scores_codex":[0.9969453,0.0001755389,0.0006283426,0.001247538,0.0002512004,0.000752127],"domain_scores_gemma":[0.9973033,0.001132417,0.0002969401,0.0008381052,0.0003147968,0.0001144285],"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":[0.00004050613,0.00007950136,0.0003815982,0.0001081477,0.00006182853,0.000001057994,0.0001185082,0.02276952,0.06212498,0.03505511,0.0004797647,0.8787795],"study_design_scores_gemma":[0.0002797339,0.0003524759,0.0003272893,0.0001662181,0.0000636214,0.000007372346,0.00004323867,0.6506365,0.2572779,0.00746436,0.0828938,0.0004875089],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004214284,0.0006093177,0.981366,0.0008481041,0.000712193,0.001987748,0.000002153222,0.0009496161,0.009310541],"genre_scores_gemma":[0.7312887,0.0002225559,0.2617912,0.0003662747,0.0001807735,0.0002805236,0.000004309253,0.00002666146,0.005838926],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.878292,"threshold_uncertainty_score":0.9998779,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0237604563867309,"score_gpt":0.3058644463728425,"score_spread":0.2821039899861116,"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."}}