{"id":"W2164441316","doi":"","title":"PORT: a Precision-Order-Recall MT Evaluation Metric for Tuning","year":2012,"lang":"en","type":"article","venue":"NPARC","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"BLEU; Computer science; Metric (unit); Port (circuit theory); Machine translation; Artificial intelligence; Precision and recall; Translation (biology); Recall; Natural language processing; Machine learning; Engineering; Electronic engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00179364,0.0001023246,0.0001166264,0.0002224491,0.0001055301,0.0001039473,0.0005985518,0.00007518238,0.0000616234],"category_scores_gemma":[0.001252526,0.00008568008,0.00004745777,0.0007731699,0.00001364025,0.0008150953,0.0001711975,0.0001086827,0.00001692952],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008013334,"about_ca_system_score_gemma":0.00007931417,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004765853,"about_ca_topic_score_gemma":0.000001055569,"domain_scores_codex":[0.9987446,0.00006187995,0.0001807654,0.0002349738,0.0004717839,0.0003059979],"domain_scores_gemma":[0.998852,0.0002278827,0.0001084181,0.0003735177,0.0003528222,0.00008542358],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000005127368,0.000047605,0.0001513968,0.0000180268,0.000007008856,6.21781e-7,0.0005583945,0.000002602839,0.01258768,0.02081452,0.004872985,0.960934],"study_design_scores_gemma":[0.0007539917,0.0001971379,0.0007592188,0.0001257339,0.00005929629,0.0000524191,0.00002138168,0.4139363,0.0634141,0.5001512,0.01993316,0.000596018],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001898298,0.002106685,0.9911904,0.0005209705,0.0003263461,0.0004290398,0.000001226658,0.0003806289,0.003146415],"genre_scores_gemma":[0.4389586,0.000003104068,0.5605894,0.0001366844,0.00008635455,0.00007197621,0.00000362026,0.000006523278,0.0001437196],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.960338,"threshold_uncertainty_score":0.3493932,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03752141005855442,"score_gpt":0.3342121167803518,"score_spread":0.2966907067217974,"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."}}