{"id":"W3177367299","doi":"10.1609/aaai.v35i16.17642","title":"Encoding Syntactic Knowledge in Transformer Encoder for Intent Detection and Slot Filling","year":2021,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"","keywords":"Computer science; Transformer; Security token; Encoder; Inference; Parsing; ENCODE; Artificial intelligence; F1 score; Benchmark (surveying); Speech recognition; Natural language processing; Pattern recognition (psychology); Voltage","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008568734,0.001278769,0.0007789076,0.001071493,0.0003307484,0.001013412,0.001759938,0.00099521,0.003724661],"category_scores_gemma":[0.002680634,0.0005944632,0.001068952,0.0007783782,0.0006323197,0.003511715,0.001210096,0.001850384,0.00265831],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007399251,"about_ca_system_score_gemma":0.001708936,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006312964,"about_ca_topic_score_gemma":0.0117275,"domain_scores_codex":[0.9996661,0.00006390001,0.00002208488,0.0001204222,0.00006764632,0.00005991189],"domain_scores_gemma":[0.9990379,0.0004444611,0.00005549998,0.0001759898,0.0002353526,0.00005091872],"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":[0.0008323939,0.0005660716,0.008116238,0.0004846568,0.0001870096,0.0005386588,0.0005998186,0.08744125,0.04385919,0.02150149,0.02391105,0.8119622],"study_design_scores_gemma":[0.00003715568,0.0001122216,0.0009316072,0.00004574687,0.0001245745,0.0002132724,0.00006517791,0.9456095,0.02579528,0.02106118,0.005962052,0.00004213953],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.054362,0.0009398829,0.9192593,0.0006301166,0.0003461958,0.0001708162,0.001897315,0.01626267,0.006131663],"genre_scores_gemma":[0.7286578,0.0007285285,0.2547856,0.0006399609,0.0001486518,0.0002577242,0.00509586,0.0006541747,0.009031758],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006312964,"threshold_uncertainty_score":0.01255244,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.103344581864608,"score_gpt":0.3068403073865154,"score_spread":0.2034957255219074,"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."}}