{"id":"W6910343010","doi":"10.48448/xmkb-qx62","title":"LLM-Based Robust Product Classification in Commerce and Compliance","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Thomson Reuters (Canada)","funders":"","keywords":"Robustness (evolution); Product (mathematics); Process (computing); Task (project management); New product development; Training set","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001456834,0.0004012194,0.0003884507,0.001696653,0.000109812,0.0002926051,0.0008594019,0.0001504769,0.0002508592],"category_scores_gemma":[0.000221055,0.0003758587,0.00002990802,0.002939719,0.002518692,0.0001961216,0.0001774713,0.0005701931,0.004501353],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004465369,"about_ca_system_score_gemma":0.0007015492,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004772077,"about_ca_topic_score_gemma":0.002571977,"domain_scores_codex":[0.9966128,0.00008644502,0.0003846226,0.001507519,0.0008256899,0.0005828969],"domain_scores_gemma":[0.9984133,0.00004946339,0.0002674086,0.000984795,0.000119307,0.0001657551],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00003365111,0.0007551534,0.002819799,0.001140848,0.00002676537,0.00004083823,0.000261621,0.0009274444,0.01720114,0.01376506,0.9442983,0.01872933],"study_design_scores_gemma":[0.001086742,0.0001466224,0.009361962,0.003505812,0.00007674029,0.00003006833,0.0003135266,0.3876691,0.0003835513,0.001038015,0.5948502,0.001537707],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.001064579,0.008560382,0.004770387,0.008951413,0.002374003,0.003796177,0.0007598578,0.002548178,0.967175],"genre_scores_gemma":[0.3349285,0.0001376295,0.07291794,0.001212612,0.001028239,0.0002191617,0.0004560562,0.002921364,0.5861785],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.3867416,"threshold_uncertainty_score":0.9998693,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1362994879092696,"score_gpt":0.3538841006291326,"score_spread":0.217584612719863,"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."}}