{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001810879,0.0001292597,0.0001656707,0.000186394,0.0004194958,0.0003120125,0.0001054873,0.0001010207,0.0000787954],"category_scores_gemma":[0.00006623629,0.0001022654,0.00006396721,0.000925242,0.00004704433,0.0001246682,0.00003312755,0.0002904944,0.00003643028],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001087695,"about_ca_system_score_gemma":0.000037563,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001131025,"about_ca_topic_score_gemma":0.00007993429,"domain_scores_codex":[0.9983951,0.00006713193,0.0005252206,0.0003724447,0.0003774466,0.0002626832],"domain_scores_gemma":[0.999262,0.00003509294,0.00008473812,0.0005313614,0.00006219395,0.00002459266],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00002970632,0.0003316097,0.005502573,0.0003999571,0.0001667438,0.0009801789,0.004915439,0.1577144,0.4504713,0.006941755,0.352597,0.01994942],"study_design_scores_gemma":[0.001087281,0.00009967804,0.0023804,0.0004469273,0.00005211406,0.0001608569,0.002617812,0.003139235,0.04868461,0.00998028,0.9307711,0.0005797707],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8946708,0.0002496177,0.001266768,0.0001127765,0.01355676,0.0004176305,4.315067e-7,0.0002433854,0.08948179],"genre_scores_gemma":[0.9848707,0.000002478242,0.00003621781,0.00003596393,0.0000888467,0.00003698161,0.00001081046,0.00001198627,0.01490602],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5781741,"threshold_uncertainty_score":0.4170263,"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."}}