{"id":"W3160270149","doi":"10.48550/arxiv.2105.04021","title":"MS MARCO: Benchmarking Ranking Models in the Large-Data Regime","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; Microsoft (Canada)","funders":"","keywords":"Benchmarking; Clef; Computer science; Ranking (information retrieval); Field (mathematics); Data science; Best practice; Track (disk drive); Artificial intelligence; Information retrieval; Political science; Task (project management); Engineering","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":[0.08415675,0.002618165,0.001838701,0.005428941,0.002273242,0.00660628,0.004524743,0.003406803,0.005942962],"category_scores_gemma":[0.1860921,0.0008100707,0.001300214,0.007097111,0.002156612,0.006944384,0.004832654,0.00457399,0.003905065],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004182185,"about_ca_system_score_gemma":0.005858833,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02464445,"about_ca_topic_score_gemma":0.04534735,"domain_scores_codex":[0.9430641,0.03817303,0.002543136,0.004143544,0.01036601,0.00171028],"domain_scores_gemma":[0.890456,0.04934315,0.003880472,0.03169935,0.01995057,0.004670492],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002049316,0.00176197,0.02826909,0.001690813,0.001732012,0.0002654913,0.0007450285,0.1610947,0.002753972,0.04527359,0.5466376,0.2077264],"study_design_scores_gemma":[0.001177864,0.001513726,0.02009148,0.000509905,0.0002578662,0.0002730499,0.0007087124,0.7844341,0.008841296,0.06144595,0.1205254,0.0002206803],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2481892,0.01716645,0.4642757,0.03124059,0.009065277,0.002621134,0.05296258,0.06815878,0.1063203],"genre_scores_gemma":[0.5758967,0.001747054,0.3025911,0.003424333,0.001516202,0.001331965,0.09412598,0.005825504,0.01354117],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.08415675,"threshold_uncertainty_score":0.4450688,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4743132750901781,"score_gpt":0.3087619711037293,"score_spread":0.1655513039864488,"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."}}