{"id":"W4407133399","doi":"10.1038/s41598-025-87046-z","title":"Learning under label noise through few-shot human-in-the-loop refinement","year":2025,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Industrial Vision Systems and Defect Detection","field":"Engineering","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Computer science; Shot (pellet); Noise (video); Loop (graph theory); Human-in-the-loop; Shot noise; Artificial intelligence; Mathematics; Telecommunications; Chemistry; Image (mathematics); Detector","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":[0.007687789,0.003540836,0.003747691,0.001399902,0.001233034,0.002245929,0.005923812,0.004401356,0.002213724],"category_scores_gemma":[0.01913012,0.001375541,0.002494696,0.001128107,0.002646674,0.004230449,0.003229832,0.005823103,0.002170094],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002074152,"about_ca_system_score_gemma":0.00197355,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0148278,"about_ca_topic_score_gemma":0.01841212,"domain_scores_codex":[0.9955751,0.001586652,0.000150705,0.001753904,0.0006232071,0.000310433],"domain_scores_gemma":[0.988879,0.007218231,0.0007451302,0.001650262,0.001106695,0.0004006565],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009275796,0.0006640332,0.005085144,0.000527616,0.0003828504,0.0004328642,0.0007870919,0.5456672,0.01460645,0.007322136,0.01145258,0.4121443],"study_design_scores_gemma":[0.0000235229,0.00009096341,0.0001932886,0.000020073,0.00001924013,0.00006937722,0.00003005932,0.9915771,0.00203528,0.005295618,0.0006278154,0.00001748039],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02658503,0.00114082,0.9649526,0.0004584101,0.0001303931,0.0001480495,0.0001718171,0.005111289,0.001301631],"genre_scores_gemma":[0.5026866,0.000755657,0.4821971,0.002135551,0.0003896639,0.0003296068,0.002455306,0.001407712,0.007642758],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0148278,"threshold_uncertainty_score":0.0406574,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04427816149620249,"score_gpt":0.3007354987626807,"score_spread":0.2564573372664782,"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."}}