{"id":"W3206689743","doi":"10.1609/aaai.v36i10.21323","title":"Non-autoregressive Translation with Layer-Wise Prediction and Deep Supervision","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"DeepMind; Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Machine translation; Autoregressive model; Inference; Computer science; Transformer; Translation (biology); Artificial intelligence; BLEU; Machine learning; Artificial neural network; Econometrics; Mathematics; Engineering; Voltage","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":[0.0003661396,0.0001730346,0.0001694296,0.0001424854,0.0004126839,0.0001884644,0.001132743,0.00005218099,0.00002924696],"category_scores_gemma":[0.00005497689,0.0001205842,0.00004319539,0.0005232533,0.0001575962,0.0005803279,0.0003715612,0.0003721328,0.000002374991],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004967442,"about_ca_system_score_gemma":0.00006200511,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003068133,"about_ca_topic_score_gemma":0.000006820943,"domain_scores_codex":[0.9984608,0.00001767011,0.000296329,0.0004415975,0.0005741649,0.0002094193],"domain_scores_gemma":[0.999101,0.00004686971,0.0002582395,0.0002059006,0.0003327931,0.00005521265],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002350013,0.0001774294,0.0008208593,0.00008595698,0.0000186933,0.000002655172,0.008588866,0.0001181775,0.1317109,0.3611684,0.00006079463,0.4970122],"study_design_scores_gemma":[0.00006006225,0.000728039,0.0002227308,0.0001997177,0.00001806744,0.00003249047,0.0008077551,0.2348996,0.6160992,0.1465991,0.00009281414,0.0002404526],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5824621,0.001146523,0.3981887,0.01031433,0.0007001368,0.002076473,0.00003053554,0.0008766889,0.004204463],"genre_scores_gemma":[0.9718595,0.00002230808,0.02786349,0.0001148518,0.00002675521,0.00006196936,0.000001120415,0.00001086577,0.00003911265],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4967718,"threshold_uncertainty_score":0.4917281,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0423277031369836,"score_gpt":0.2771226150452021,"score_spread":0.2347949119082186,"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."}}