{"id":"W3156861296","doi":"10.1145/3404835.3463034","title":"Significant Improvements over the State of the Art? A Case Study of the MS MARCO Document Ranking Leaderboard","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Microsoft (Canada); University of Waterloo","funders":"","keywords":"Ranking (information retrieval); Metric (unit); Context (archaeology); Computer science; State (computer science); sort; Artificial intelligence; Information retrieval; Order (exchange); Machine learning; Algorithm; Engineering; Geography","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":[],"consensus_categories":[],"category_scores_codex":[0.0005244813,0.0001076008,0.0001438788,0.00001783331,0.0001418044,0.00007947792,0.001004241,0.00001496776,0.00002785198],"category_scores_gemma":[0.00002583416,0.00004701469,0.00009556754,0.0002991299,0.00005088802,0.0001175468,0.001129598,0.0001434113,0.000001612687],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003287363,"about_ca_system_score_gemma":0.0001046646,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001445429,"about_ca_topic_score_gemma":0.0008386411,"domain_scores_codex":[0.9983134,0.0002482122,0.0003751251,0.0002890673,0.0005764381,0.0001978116],"domain_scores_gemma":[0.9982072,0.0001015572,0.0001795824,0.001398587,0.00008925801,0.00002380957],"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":[0.00008781974,0.005019384,0.06579041,0.0004533605,0.001894737,0.001668714,0.2684936,0.06348059,0.1613483,0.0439123,0.003385436,0.3844654],"study_design_scores_gemma":[0.01124812,0.0009784362,0.06216155,0.0004436121,0.000369321,0.000548149,0.05409604,0.5461312,0.289574,0.03005303,0.003009236,0.001387341],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.964211,0.00002837727,0.0331547,0.001040894,0.0003703182,0.0005932683,0.000001168965,0.0000147224,0.0005855726],"genre_scores_gemma":[0.9969999,0.000002292419,0.0008119908,0.0003999116,0.0000159465,0.00001812466,5.316885e-8,0.00000546105,0.001746295],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4826506,"threshold_uncertainty_score":0.2185066,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02407949558039306,"score_gpt":0.2593646241959989,"score_spread":0.2352851286156058,"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."}}