{"id":"W3120179416","doi":"10.18653/v1/2020.wmt-1.99","title":"Extended Study on Using Pretrained Language Models and YiSi-1 for Machine Translation Evaluation","year":2020,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"Machine translation; Computer science; Task (project management); Artificial intelligence; Natural language processing; Translation (biology); Evaluation of machine translation; BLEU; Language model; Quality (philosophy); Embedding; Machine learning; Example-based machine translation; Machine translation software usability","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01446701,0.002666675,0.001246984,0.002063847,0.0006555152,0.002088697,0.001752071,0.001325779,0.003545596],"category_scores_gemma":[0.05017663,0.0006313009,0.0009861684,0.002424644,0.0008125327,0.004399806,0.002101733,0.003615268,0.00183628],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001413623,"about_ca_system_score_gemma":0.001995466,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008055374,"about_ca_topic_score_gemma":0.01330883,"domain_scores_codex":[0.9910424,0.004551359,0.0007426015,0.001487733,0.00176908,0.0004067929],"domain_scores_gemma":[0.9512072,0.0242134,0.001959222,0.01063405,0.01127219,0.0007139447],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001864048,0.001734796,0.04848801,0.001064736,0.00175974,0.0003532929,0.0005927352,0.1999599,0.0355341,0.004331681,0.02087765,0.6834393],"study_design_scores_gemma":[0.00008377364,0.002247755,0.02285117,0.0001601254,0.0002914939,0.0003477189,0.000191249,0.9108247,0.0520087,0.004150748,0.006691128,0.0001514755],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6181668,0.009888154,0.3334029,0.00184283,0.000989196,0.0006886724,0.004267517,0.01265559,0.01809838],"genre_scores_gemma":[0.8136781,0.001323223,0.1641081,0.0005308018,0.0002634291,0.0004302369,0.01283623,0.0009994254,0.005830479],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01446701,"threshold_uncertainty_score":0.07650971,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1102848588545824,"score_gpt":0.3752472339290164,"score_spread":0.264962375074434,"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."}}