{"id":"W3154476213","doi":"10.18653/v1/2021.cmcl-1.9","title":"TorontoCL at CMCL 2021 Shared Task: RoBERTa with Multi-Stage Fine-Tuning for Eye-Tracking Prediction","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"Canadian Institute for Advanced Research","keywords":"Computer science; Transformer; Eye tracking; Task (project management); Ranking (information retrieval); Artificial intelligence; Comprehension; Language model; Natural language processing; Tracking (education); Reading comprehension; Task analysis; Machine learning; Reading (process); Programming language; Engineering","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":[0.000190469,0.0001714542,0.0001835407,0.00004025287,0.0002415863,0.0002581954,0.0004128857,0.00007934653,0.0002394317],"category_scores_gemma":[0.00009006816,0.0001512224,0.00007210033,0.0001612593,0.00001347148,0.0008272285,0.0003355443,0.0001031927,0.00001331644],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001640616,"about_ca_system_score_gemma":0.0001085882,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002068379,"about_ca_topic_score_gemma":0.003283217,"domain_scores_codex":[0.9983903,0.0000368098,0.0002680358,0.0006758347,0.0002597952,0.0003692895],"domain_scores_gemma":[0.9987915,0.00009069232,0.00008730501,0.0006936581,0.0002314159,0.0001054157],"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":[0.0002274766,0.0009225812,0.1094059,0.0007486885,0.0006311362,0.0005782971,0.02567133,0.2141156,0.2778673,0.04373959,0.01115137,0.3149407],"study_design_scores_gemma":[0.0008948159,0.00004848272,0.002414548,0.00007486727,0.00001092127,0.00001311082,0.0001210995,0.9820335,0.006529541,0.00003021163,0.007616099,0.0002128225],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01449015,0.000204863,0.9824331,0.0005563084,0.0003749597,0.0002735801,0.00001404283,0.0001782862,0.00147468],"genre_scores_gemma":[0.3617451,0.000006984097,0.6107921,0.0002291711,0.0001947904,0.00007195637,0.00005652832,0.0000258093,0.02687754],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7679178,"threshold_uncertainty_score":0.616667,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04593062006633086,"score_gpt":0.2917247730476185,"score_spread":0.2457941529812876,"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."}}