{"id":"W4389524312","doi":"10.18653/v1/2023.nllp-1.26","title":"Tracing Influence at Scale: A Contrastive Learning Approach to Linking Public Comments and Regulator Responses","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ivey Foundation; University of British Columbia","funders":"Social Sciences and Humanities Research Council of Canada","keywords":"Tracing; Matching (statistics); Computer science; Set (abstract data type); Scale (ratio); Regulator; Test set; Test (biology); Simple (philosophy); Artificial intelligence; Machine learning; Programming language; Mathematics; Epistemology","routes":{"ca_aff":true,"ca_fund":true,"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":[0.004522301,0.00115511,0.0007980178,0.003755203,0.0009479094,0.001687819,0.002287539,0.002195025,0.001808263],"category_scores_gemma":[0.02297388,0.000590843,0.001001255,0.002040117,0.001555503,0.003281907,0.002615308,0.002580283,0.0009487944],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001770263,"about_ca_system_score_gemma":0.001014089,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009610169,"about_ca_topic_score_gemma":0.01382802,"domain_scores_codex":[0.9966449,0.001701712,0.0001243186,0.0009954995,0.0003828254,0.0001508669],"domain_scores_gemma":[0.9815379,0.01457503,0.001078667,0.00124737,0.001300143,0.0002608075],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001132438,0.0009749691,0.04486513,0.0004332212,0.0005452858,0.0009619751,0.004631494,0.1935422,0.02811121,0.01977994,0.01022135,0.6948007],"study_design_scores_gemma":[0.00003720427,0.0000934724,0.003435937,0.00001885665,0.00005507955,0.00009004279,0.0001914623,0.9781851,0.004461203,0.01108682,0.002316605,0.00002827352],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2214394,0.0009379279,0.7635882,0.001412505,0.0001564777,0.000418388,0.00071547,0.003985942,0.007345723],"genre_scores_gemma":[0.8655474,0.0001653239,0.1285441,0.0004358305,0.0002014839,0.0002333472,0.001057912,0.0002027087,0.003611978],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009610169,"threshold_uncertainty_score":0.02391654,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03797563552236556,"score_gpt":0.2654466282762636,"score_spread":0.2274709927538981,"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."}}