{"id":"W4298095693","doi":"10.48550/arxiv.2109.10739","title":"Predicting Efficiency/Effectiveness Trade-offs for Dense vs. Sparse\\n Retrieval Strategy Selection","year":2021,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Leverage (statistics); Embedding; Classifier (UML); Artificial intelligence","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"],"consensus_categories":[],"category_scores_codex":[0.002197194,0.001024506,0.001182402,0.0005597558,0.001159672,0.0008011387,0.002463747,0.00110527,0.00005228559],"category_scores_gemma":[0.0003968155,0.001372908,0.000884056,0.002455464,0.0002614755,0.001245592,0.001648844,0.001694652,0.00001718197],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001364863,"about_ca_system_score_gemma":0.001963907,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003522505,"about_ca_topic_score_gemma":0.0001425772,"domain_scores_codex":[0.9916868,0.001167708,0.0008572132,0.004437676,0.0004003228,0.001450235],"domain_scores_gemma":[0.9949691,0.001119373,0.0008631761,0.001783661,0.0007487578,0.0005159015],"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":[0.0006578368,0.0003784845,0.01370807,0.0007774691,0.0002864302,0.0003046949,0.0008795837,0.9610932,0.0009328116,0.01907294,0.000008243387,0.001900193],"study_design_scores_gemma":[0.002017295,0.0006190064,0.007978014,0.0007044802,0.0004127833,0.00007657994,0.0005012788,0.9783049,0.004745028,0.00338896,0.00004793339,0.001203767],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4805949,0.0001387873,0.5160252,0.00003597057,0.001680721,0.001111521,0.00001581634,0.0002273786,0.0001697395],"genre_scores_gemma":[0.9960148,0.0001834665,0.002666434,0.00005202456,0.0004952777,0.000005753996,0.00003243156,0.00007596035,0.0004738542],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5154199,"threshold_uncertainty_score":0.998872,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08046011134342035,"score_gpt":0.2106495405814781,"score_spread":0.1301894292380577,"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."}}