{"id":"W7039767753","doi":"","title":"Natural Language Reasoning with Transformer Language Models","year":2023,"lang":"fr","type":"other","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Environmental Monitoring and Data Management","field":"Earth and Planetary Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Transformer; Natural language understanding; Natural language","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0005361002,0.0008260791,0.0006402865,0.000497659,0.0003379862,0.0003380753,0.000830107,0.0004855385,0.002791382],"category_scores_gemma":[0.00004182565,0.0007094596,0.0002189009,0.0005799941,0.0002428834,0.0007070679,0.00008596427,0.0009915035,0.0009011725],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002614252,"about_ca_system_score_gemma":0.00009851003,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.1674185,"about_ca_topic_score_gemma":0.1784015,"domain_scores_codex":[0.9962829,0.0001685237,0.000471414,0.0009671889,0.0007572074,0.00135279],"domain_scores_gemma":[0.9980323,0.0001517458,0.0002657747,0.001063661,0.00001585796,0.0004706564],"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":[0.0005534532,0.0003988575,0.1767257,0.0007555856,0.001078894,0.004540257,0.007556913,0.08102012,0.0009007109,0.005970734,0.01291363,0.7075852],"study_design_scores_gemma":[0.002840577,0.001290022,0.3915679,0.003747229,0.0009803079,0.0006019959,0.01134042,0.4760792,0.002349641,0.0002786025,0.1038656,0.005058563],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.2747506,0.1322741,0.1925653,0.00687255,0.00496671,0.007344629,0.005794031,0.009193368,0.3662387],"genre_scores_gemma":[0.5253527,0.003233026,0.03411912,0.000847987,0.0008384632,0.0001572108,0.0008794357,0.0005326818,0.4340394],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.7025266,"threshold_uncertainty_score":0.9998767,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007871666646805255,"score_gpt":0.2010996627027025,"score_spread":0.1932279960558972,"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."}}