{"id":"W4400681662","doi":"10.1109/saner60148.2024.00019","title":"Refining GPT-3 Embeddings with a Siamese Structure for Technical Post Duplicate Detection","year":2024,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Refining (metallurgy); Computer science; Artificial intelligence; Materials science; Metallurgy","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":[0.001306063,0.00161552,0.0009791944,0.002136612,0.0006322369,0.0011108,0.001396632,0.001532993,0.002225529],"category_scores_gemma":[0.007126869,0.0004808222,0.001391345,0.001907103,0.001109083,0.003135887,0.001754226,0.002227188,0.0019358],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009552435,"about_ca_system_score_gemma":0.001816155,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00759154,"about_ca_topic_score_gemma":0.01219863,"domain_scores_codex":[0.999128,0.0002020002,0.00006508173,0.0003053776,0.0002098877,0.00008970981],"domain_scores_gemma":[0.9976462,0.0006790174,0.000352644,0.000491058,0.0006924903,0.0001386247],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0005038019,0.0006083752,0.017394,0.000431266,0.0002766807,0.0008243611,0.0006684798,0.3091692,0.02674808,0.01835198,0.02259561,0.6024283],"study_design_scores_gemma":[0.00001793732,0.00008851393,0.001114846,0.00001536847,0.00002406642,0.0001926007,0.00008014501,0.9833387,0.00320234,0.009765109,0.002135872,0.00002451503],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1338602,0.0006856697,0.8562491,0.0005164176,0.000158146,0.0002461698,0.001278512,0.004205104,0.00280069],"genre_scores_gemma":[0.6582815,0.0005439784,0.3254167,0.0004132118,0.0001537687,0.0004369968,0.007355277,0.0005009314,0.006897562],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00759154,"threshold_uncertainty_score":0.0150947,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007164724710223737,"score_gpt":0.2590984125606616,"score_spread":0.2519336878504378,"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."}}