{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002796946,0.0007292037,0.001114009,0.001513666,0.0006412297,0.00160468,0.001434081,0.001393578,0.001753763],"category_scores_gemma":[0.009684941,0.0003690657,0.0008358944,0.00100736,0.001013436,0.002145912,0.001730679,0.001381369,0.001254777],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001430859,"about_ca_system_score_gemma":0.001223354,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005350174,"about_ca_topic_score_gemma":0.004365863,"domain_scores_codex":[0.9981532,0.0006724441,0.0001329729,0.0004802646,0.0003973918,0.0001636692],"domain_scores_gemma":[0.9948587,0.002219676,0.0007527397,0.001244981,0.0007770624,0.0001468799],"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":[0.0005931824,0.0004226058,0.01999295,0.0001799941,0.0001170252,0.000314818,0.0001633345,0.5713843,0.005800789,0.01192542,0.00714075,0.3819648],"study_design_scores_gemma":[0.00000421347,0.00001652925,0.0006800961,0.000005255858,0.000003355585,0.00001875472,0.00001177569,0.9944677,0.0009687192,0.003392196,0.0004262364,0.000005049499],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.183177,0.0009762767,0.8028585,0.001369926,0.0001772617,0.0001745472,0.0008139519,0.004854485,0.005597972],"genre_scores_gemma":[0.8609018,0.0001428134,0.1344481,0.0002788858,0.00008605683,0.00008661491,0.001352433,0.0001475188,0.002555863],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005350174,"threshold_uncertainty_score":0.01479185,"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."}}