{"id":"W4376561185","doi":"10.1016/j.compind.2023.103939","title":"Full-cycle data purification strategy for multi-type weld seam classification with few-shot learning","year":2023,"lang":"en","type":"article","venue":"Computers in Industry","topic":"Welding Techniques and Residual Stresses","field":"Engineering","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Welding; Computer science; Artificial intelligence; Shot (pellet); Key (lock); Randomness; Volume (thermodynamics); Pattern recognition (psychology); Data mining; Computer vision; Engineering; Mathematics; Mechanical engineering","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":[0.0001609504,0.0001234842,0.0001166353,0.000121158,0.00006743185,0.00006956796,0.0003767162,0.0002745267,0.000006720079],"category_scores_gemma":[0.00002927537,0.0001181028,0.00001283144,0.000437276,0.0000276933,0.0001486429,0.00007522926,0.0004958652,0.000009216793],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005199415,"about_ca_system_score_gemma":0.00002164122,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001787947,"about_ca_topic_score_gemma":0.00001953049,"domain_scores_codex":[0.9992161,0.0000204213,0.0001745377,0.0002648941,0.00009756091,0.0002264912],"domain_scores_gemma":[0.9994299,0.00007678466,0.00004060076,0.0003737944,0.0000329256,0.00004602241],"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":[0.0000468842,0.00007704736,0.005038749,0.0002039828,0.00004906455,0.00001329535,0.0001679844,0.9275162,0.00503106,0.0002479597,0.02662699,0.03498083],"study_design_scores_gemma":[0.0003916152,0.0001664018,0.01469524,0.0001927301,0.00001089186,0.000005922493,0.0003285303,0.9737248,0.003737956,0.00008264738,0.00639926,0.0002639613],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8548689,0.0001620225,0.1398158,0.0003877658,0.0006884481,0.0007551527,0.00004018684,0.002400066,0.0008815719],"genre_scores_gemma":[0.9836968,0.00006621875,0.01552199,0.000009740839,0.0001396286,0.00003906091,0.00040631,0.00003739927,0.00008289196],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1288278,"threshold_uncertainty_score":0.4816093,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2035916618725392,"score_gpt":0.3492681815331888,"score_spread":0.1456765196606497,"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."}}