{"id":"W4385572760","doi":"10.18653/v1/2022.emnlp-main.23","title":"Certified Error Control of Candidate Set Pruning for Two-Stage Relevance Ranking","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund; Compute Canada","keywords":"Pruning; Ranking (information retrieval); Computer science; Relevance (law); Set (abstract data type); Data mining; Domain (mathematical analysis); Error detection and correction; Artificial intelligence; Machine learning; Algorithm; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01942173,0.001650534,0.002414646,0.002605051,0.001589862,0.003337181,0.004198875,0.002605699,0.00259075],"category_scores_gemma":[0.1216019,0.0008375843,0.001431707,0.0019006,0.002878963,0.00390717,0.00373626,0.003208329,0.001703488],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001712441,"about_ca_system_score_gemma":0.004451521,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003517769,"about_ca_topic_score_gemma":0.004468946,"domain_scores_codex":[0.9730728,0.009627005,0.002315853,0.004083341,0.009513223,0.001387723],"domain_scores_gemma":[0.8846298,0.06882907,0.00681504,0.02375657,0.01453675,0.001432832],"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":[0.002533982,0.000678383,0.0156106,0.0007398037,0.0002923599,0.0004822412,0.0007449038,0.2723601,0.04176629,0.04325218,0.01527058,0.6062686],"study_design_scores_gemma":[0.000224277,0.0004931947,0.002160999,0.00009327633,0.00008142009,0.0004913482,0.00007738914,0.9460673,0.02356281,0.02274234,0.003922387,0.00008317368],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03592713,0.0009015926,0.9552848,0.0003589488,0.0001732858,0.0003544322,0.0002175341,0.004415367,0.00236697],"genre_scores_gemma":[0.5582587,0.0002544319,0.4355142,0.0005148759,0.0002589747,0.0006964101,0.00110817,0.001024675,0.002369589],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01942173,"threshold_uncertainty_score":0.1027131,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04368542010145735,"score_gpt":0.2910072447144538,"score_spread":0.2473218246129965,"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."}}