{"id":"W2597891111","doi":"10.1016/j.csl.2017.01.014","title":"On integrating a language model into neural machine translation","year":2017,"lang":"en","type":"article","venue":"Computer Speech & Language","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":113,"is_retracted":false,"has_abstract":false,"ca_institutions":"Canadian Institute for Advanced Research; Université de Montréal","funders":"","keywords":"Machine translation; Computer science; Artificial intelligence; Phrase; Natural language processing; Translation (biology); BLEU; Language model; Baseline (sea); Example-based machine translation; Turkish; Machine learning; Linguistics","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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003467765,0.0003438322,0.0003028657,0.0002001906,0.0005193943,0.001087925,0.002866655,0.0001380374,0.00001731672],"category_scores_gemma":[0.00009582692,0.0002844595,0.0001408179,0.0001436488,0.00007586322,0.00111774,0.0006095375,0.0005497884,0.00003158267],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007442103,"about_ca_system_score_gemma":0.00004194971,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003420242,"about_ca_topic_score_gemma":0.0001782415,"domain_scores_codex":[0.9981158,0.00007171216,0.0003077427,0.000672801,0.0004254754,0.0004064203],"domain_scores_gemma":[0.9976716,0.00007851947,0.0002650729,0.001784705,0.0000710696,0.0001290162],"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":[0.00001172026,0.00004993957,0.00002480923,0.00002926339,0.00001351145,0.0003124067,0.008423117,0.0001385726,0.005458472,0.009693504,0.0003425785,0.9755021],"study_design_scores_gemma":[0.0003801496,0.0001087894,0.0000546902,0.00009052255,0.000008830203,0.00005263875,0.00003255042,0.9754,0.01612655,0.007355284,0.00003077625,0.0003592676],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03297767,0.002511402,0.9596383,0.001367731,0.0003723477,0.0002657454,0.000006773487,0.001197676,0.00166236],"genre_scores_gemma":[0.5062477,0.00000200406,0.4928146,0.0006334964,0.0001481502,0.000007986948,0.00001076548,0.00001989694,0.0001154352],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9752614,"threshold_uncertainty_score":0.9999608,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01622676902070584,"score_gpt":0.3056925943937503,"score_spread":0.2894658253730444,"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."}}