{"id":"W2791176029","doi":"10.1016/j.neucom.2018.01.007","title":"Fine-grained attention mechanism for neural machine translation","year":2018,"lang":"en","type":"preprint","venue":"Neurocomputing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; National Research Foundation of Korea; National Research Foundation; Compute Canada; Samsung; Canada Research Chairs; Canadian Institute for Advanced Research","keywords":"Machine translation; Computer science; Mechanism (biology); Translation (biology); Artificial intelligence; Context (archaeology); Word (group theory); Natural language processing; Exploit; Task (project management); Machine learning; Linguistics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004531765,0.000405636,0.0003595779,0.0002578553,0.0002837556,0.0004800626,0.001590388,0.0003009934,0.000002359607],"category_scores_gemma":[0.0001275016,0.0004007204,0.000267396,0.000248055,0.00003090867,0.0002641004,0.001075421,0.0006621325,0.000004605353],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005073193,"about_ca_system_score_gemma":0.00006118492,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002293962,"about_ca_topic_score_gemma":0.00001025851,"domain_scores_codex":[0.997495,0.0001164061,0.0005150013,0.001077965,0.0003528496,0.0004427792],"domain_scores_gemma":[0.9982071,0.0001803563,0.0004755489,0.0007179915,0.0003417561,0.00007723409],"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":[0.00004397257,0.0001092164,0.00007728712,0.001105944,0.00005277064,0.00003638949,0.0007477879,0.0005885742,0.02421391,0.04286524,0.0009659827,0.9291929],"study_design_scores_gemma":[0.0002824987,0.0001102647,0.0000391597,0.0001591052,0.00002136915,0.00002371337,8.115379e-7,0.8777261,0.005849861,0.1153242,0.0001085694,0.0003543272],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008153594,0.0006289872,0.9846614,0.001829864,0.001527035,0.001045138,0.00001439212,0.002063967,0.00007560607],"genre_scores_gemma":[0.4483984,0.000001576063,0.5506247,0.0003348856,0.0004706492,0.00004890955,0.00005741175,0.0000361127,0.0000273385],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9288386,"threshold_uncertainty_score":0.9998445,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03056099108739277,"score_gpt":0.29747321397155,"score_spread":0.2669122228841573,"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."}}