{"id":"W3028435993","doi":"10.5220/0009422400580067","title":"Using BERT and XLNET for the Automatic Short Answer Grading Task","year":2020,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":46,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Task (project management); Grading (engineering); Artificial intelligence; Natural language processing; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000129838,0.00005973628,0.00007194519,0.00001288527,0.0001177648,0.0001274173,0.0002995323,0.00001798082,0.000006907661],"category_scores_gemma":[0.00002829263,0.00003914869,0.0000239806,0.00008346371,0.00001354664,0.0001838094,0.0001348264,0.00004184109,0.000001388146],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000008424929,"about_ca_system_score_gemma":0.00001694166,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002038393,"about_ca_topic_score_gemma":0.000003455808,"domain_scores_codex":[0.999465,0.00001056086,0.0001071332,0.0001941217,0.00009195897,0.0001311521],"domain_scores_gemma":[0.9996215,0.000114065,0.00001385447,0.0001846538,0.00001443798,0.00005150491],"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":[0.000005683692,0.00003155101,0.006821745,0.0002088054,0.0001348763,0.00001310458,0.01433054,0.01219054,0.01160195,0.2990274,0.005351734,0.6502821],"study_design_scores_gemma":[0.00005706576,0.00001152951,0.0001828768,0.000005235163,0.000006727855,0.000007040136,0.00003788099,0.9971176,0.0001785652,0.001135973,0.001200091,0.00005945138],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02582414,0.00008614573,0.9690167,0.004360179,0.0001198498,0.0001682635,3.490317e-7,0.0001055993,0.0003187637],"genre_scores_gemma":[0.7114656,0.000001677252,0.2869393,0.001488786,0.00006888402,0.000004647982,1.25008e-7,0.000003715237,0.00002723235],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.984927,"threshold_uncertainty_score":0.1596437,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1021931580915548,"score_gpt":0.294745850003279,"score_spread":0.1925526919117242,"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."}}