{"id":"W4397012280","doi":"10.1017/nlp.2024.7","title":"A survey of context in neural machine translation and its evaluation","year":2024,"lang":"en","type":"article","venue":"Natural language processing.","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"Dublin City University; Science Foundation Ireland","keywords":"Machine translation; Computer science; Artificial intelligence; Evaluation of machine translation; Context (archaeology); Natural language processing; Terminology; Machine translation software usability; Consistency (knowledge bases); Example-based machine translation; Sentence; Computer-assisted translation; Task (project management); Paragraph; Transfer-based machine translation; Machine learning; Linguistics; World Wide Web; Engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"simulation_or_modeling","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"design_other","genre":"review","about_ca_system":false,"about_ca_topic":false,"confidence":"high","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01459207,0.001505316,0.002215877,0.005426799,0.0009528288,0.002986988,0.00232088,0.002165894,0.00245355],"category_scores_gemma":[0.0428756,0.0005654492,0.001071757,0.004938262,0.001059618,0.003304939,0.001921739,0.001297651,0.0006592994],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002650352,"about_ca_system_score_gemma":0.001526525,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006481652,"about_ca_topic_score_gemma":0.00597721,"domain_scores_codex":[0.9810293,0.01208333,0.001605879,0.00158278,0.00338028,0.0003183962],"domain_scores_gemma":[0.9710484,0.01944213,0.001335664,0.001746735,0.006013337,0.0004139499],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001611747,0.0002274243,0.008395502,0.003897435,0.0009233701,0.00009182491,0.0002090971,0.043231,0.002981775,0.005636111,0.005521835,0.9272729],"study_design_scores_gemma":[0.0005812291,0.004859946,0.03096424,0.00749666,0.002693787,0.001068671,0.001099785,0.7670239,0.04423236,0.05328752,0.08629181,0.0004000997],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"review","genre_gemma":"empirical","genre_scores_codex":[0.1701156,0.6623136,0.1246416,0.003520195,0.0009076901,0.0005511797,0.001355852,0.002970085,0.03362423],"genre_scores_gemma":[0.8122588,0.06769282,0.1117437,0.001003793,0.000789605,0.0004311239,0.002638011,0.0006064857,0.002835674],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01459207,"threshold_uncertainty_score":0.07717115,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03055695989947995,"score_gpt":0.3326759585021715,"score_spread":0.3021189986026915,"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."}}