{"id":"W2531207078","doi":"10.1162/tacl_a_00067","title":"Fully Character-Level Neural Machine Translation without Explicit Segmentation","year":2017,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":415,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"York University; Samsung Advanced Institute of Technology; Samsung; Nvidia","keywords":"Computer science; Machine translation; Character (mathematics); Pooling; Encoder; Convolutional neural network; Artificial intelligence; Translation (biology); Natural language processing; Segmentation; Speech recognition; Task (project management); Language model; Representation (politics)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004019013,0.0009828565,0.0006612442,0.0004598483,0.0003852232,0.0008666583,0.001083708,0.0008740707,0.006123744],"category_scores_gemma":[0.002153294,0.0003439613,0.0005900062,0.001021107,0.0003785697,0.001810606,0.0009074261,0.001024423,0.004576148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005063899,"about_ca_system_score_gemma":0.00116801,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003681072,"about_ca_topic_score_gemma":0.007645729,"domain_scores_codex":[0.9995779,0.00008596748,0.00003321502,0.0001669619,0.00008662008,0.00004926735],"domain_scores_gemma":[0.9991353,0.0002345918,0.0000637335,0.0003126864,0.0002257725,0.00002789007],"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":[0.0005324047,0.0003176903,0.001353153,0.000491166,0.0001610785,0.0004859472,0.0001808849,0.1845673,0.09449791,0.0183189,0.02080486,0.6782887],"study_design_scores_gemma":[0.00002489029,0.0001186049,0.0005879584,0.00002126631,0.00003288747,0.000222429,0.00002794045,0.9396688,0.03822535,0.01116515,0.009879828,0.0000249355],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06779244,0.0007854792,0.9006131,0.0003423279,0.0002369238,0.0001127313,0.001358714,0.01466913,0.014089],"genre_scores_gemma":[0.6053543,0.0005226865,0.3657224,0.0003910598,0.000104567,0.0002041542,0.007608874,0.0009344073,0.01915754],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006123744,"threshold_uncertainty_score":0.020486,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0388812594394734,"score_gpt":0.3152326302466266,"score_spread":0.2763513708071532,"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."}}