{"id":"W4375841107","doi":"10.20944/preprints202305.0421.v1","title":"An Optimized Approach to Translate Technical Patents from English to Japanese Using Machine Translation Models","year":2023,"lang":"en","type":"preprint","venue":"Preprints.org","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Engineers Without Borders Canada; Western University","funders":"","keywords":"Machine translation; Computer science; Artificial intelligence; Documentation; Evaluation of machine translation; Natural language processing; Scope (computer science); Translation (biology); Technical documentation; Field (mathematics); Machine translation software usability; Machine learning; Example-based machine translation; Programming language","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","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.004126003,0.0008942392,0.001259074,0.00048491,0.000311269,0.0003895623,0.003380952,0.0007069673,0.0009599326],"category_scores_gemma":[0.0005289684,0.0009083716,0.0002771056,0.000599643,0.0001510484,0.00061401,0.001975665,0.00106435,0.001344697],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002601626,"about_ca_system_score_gemma":0.0001848036,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007888981,"about_ca_topic_score_gemma":0.00008963612,"domain_scores_codex":[0.9915214,0.001084204,0.001430971,0.003513954,0.001398344,0.001051086],"domain_scores_gemma":[0.9953166,0.0001699017,0.0004376665,0.002972583,0.0003175858,0.0007856298],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002302891,0.0001529559,0.001609413,0.00007108444,0.00001300367,0.000005740834,0.006553907,0.5327824,0.4584947,0.00001032149,0.000002733562,0.00007349405],"study_design_scores_gemma":[0.001094757,0.00007537624,0.01523109,0.0004258715,0.0001761694,0.000007954653,0.0002069194,0.8886485,0.08992968,0.002351848,0.00007373521,0.001778097],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8756376,0.00002811357,0.1163182,0.0002131045,0.002182418,0.002416188,0.0005770132,0.001667187,0.0009602094],"genre_scores_gemma":[0.7963326,0.000009957604,0.2021575,0.000151397,0.0003655499,0.0004244121,0.000298277,0.0001713303,0.00008895779],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.368565,"threshold_uncertainty_score":0.9999533,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2231368210762454,"score_gpt":0.3802667621165625,"score_spread":0.157129941040317,"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."}}