{"id":"W4410827183","doi":"10.1007/s10845-025-02623-3","title":"Focus to generalize (F2G): physics-guided attention for sample efficient and generalizable deep learning defect detection","year":2025,"lang":"en","type":"article","venue":"Journal of Intelligent Manufacturing","topic":"Industrial Vision Systems and Defect Detection","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"National Research Council Canada; University of Toronto","funders":"","keywords":"Focus (optics); Sample (material); Artificial intelligence; Deep learning; Computer science; Cognitive science; Machine learning; Engineering; Psychology; Physics; Optics","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":[0.0008390077,0.001693228,0.001030941,0.0008331953,0.0003370129,0.0005311972,0.002268042,0.001761422,0.004885619],"category_scores_gemma":[0.00198734,0.0004755738,0.0008364971,0.0006687964,0.0005712406,0.001254991,0.002453788,0.001792823,0.0009709306],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000685788,"about_ca_system_score_gemma":0.001129098,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00659668,"about_ca_topic_score_gemma":0.01092118,"domain_scores_codex":[0.9997699,0.00003589289,0.000006955805,0.00009054202,0.00005702315,0.00003982319],"domain_scores_gemma":[0.9995363,0.0001758597,0.00003270915,0.0001219687,0.00008203976,0.00005103953],"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":[0.000381083,0.0002889405,0.001590655,0.0002829955,0.0001674393,0.0002362273,0.0001587686,0.1196891,0.03877649,0.01548677,0.0339524,0.7889891],"study_design_scores_gemma":[0.00002826005,0.0001119028,0.0004222191,0.00001270991,0.00002333976,0.0000659252,0.00001327133,0.9762368,0.008891881,0.01218239,0.00200036,0.00001093568],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04046028,0.000794457,0.9451762,0.0004984869,0.0001544651,0.0001091002,0.0005372972,0.009948581,0.002321018],"genre_scores_gemma":[0.5893378,0.0004232215,0.3963806,0.001125964,0.0001839912,0.0002452258,0.001714192,0.001219411,0.009369582],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00659668,"threshold_uncertainty_score":0.01634401,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01811174095409849,"score_gpt":0.2584816367091504,"score_spread":0.2403698957550519,"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."}}