{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007529283,0.0001757652,0.0002101061,0.0001666933,0.0004841982,0.0001669205,0.002515686,0.00003613818,0.0001206303],"category_scores_gemma":[0.0001450325,0.0001534986,0.000102852,0.0007670202,0.00005186329,0.0002443486,0.001217639,0.0005413743,0.00006456472],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007892781,"about_ca_system_score_gemma":0.00006540214,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001070901,"about_ca_topic_score_gemma":0.000003024267,"domain_scores_codex":[0.9979625,0.00003216676,0.0004352304,0.0005338431,0.0006737466,0.0003625521],"domain_scores_gemma":[0.9990546,0.00006472765,0.0002262906,0.0003184748,0.000240363,0.00009559513],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004790381,0.00009112098,0.00006010968,0.00002123936,0.00001123958,0.000001493718,0.01463821,0.2795329,0.04841625,0.5460386,0.00005187108,0.1110891],"study_design_scores_gemma":[0.00001601062,0.000134359,0.000004448393,0.00003582394,0.000004386452,0.000005610785,0.001810995,0.8469016,0.1119274,0.03875512,0.0002304312,0.0001738659],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2186244,0.00002292226,0.7724751,0.003249388,0.0004285668,0.0003418064,0.000001559037,0.000147016,0.004709275],"genre_scores_gemma":[0.9868441,0.000004075758,0.01211688,0.0003826707,0.00005897527,0.00005325406,2.544261e-7,0.00001358638,0.0005261843],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7682198,"threshold_uncertainty_score":0.625949,"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."}}