{"id":"W3207058314","doi":"10.1609/aaai.v36i10.21308","title":"Language Modelling via Learning to Rank","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Vector Institute","funders":"Natural Sciences and Engineering Research Council of Canada; Vector Institute","keywords":"Perplexity; Computer science; Language model; Transformer; Natural language processing; Artificial intelligence; Probabilistic logic; Entropy (arrow of time); Machine learning; Rank (graph theory); Ranking (information retrieval); Mathematics","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":[0.001997774,0.001308076,0.0008823154,0.001311199,0.0006316051,0.002215087,0.001919976,0.001407026,0.005689178],"category_scores_gemma":[0.009500859,0.0004862206,0.001122421,0.0009962915,0.0007764206,0.00306599,0.001835038,0.00251544,0.005738337],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001060499,"about_ca_system_score_gemma":0.001312722,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004915467,"about_ca_topic_score_gemma":0.008937541,"domain_scores_codex":[0.9981182,0.0009136575,0.00008217493,0.0004376905,0.0003063631,0.0001418636],"domain_scores_gemma":[0.9952723,0.003016482,0.0002606235,0.0007065742,0.0006019359,0.0001420862],"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":[0.0002375004,0.0001683536,0.001668028,0.0002890913,0.0001399014,0.0001752627,0.000270066,0.4676259,0.004981708,0.04590588,0.01881905,0.4597193],"study_design_scores_gemma":[0.00001324223,0.00003437788,0.0001070587,0.00001271818,0.0000103895,0.00003449124,0.00001989351,0.9590054,0.001718479,0.0369519,0.002076312,0.00001574406],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01053754,0.0004543177,0.9792991,0.0006839253,0.00008723248,0.00005734032,0.0005436781,0.00526363,0.003073329],"genre_scores_gemma":[0.5003977,0.0006993503,0.4775813,0.0007265804,0.0003367762,0.0003972575,0.00397886,0.000930877,0.01495123],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005689178,"threshold_uncertainty_score":0.01903218,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07286834451957244,"score_gpt":0.2874143963526772,"score_spread":0.2145460518331048,"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."}}