{"id":"W3185884227","doi":"10.17533/udea.mut.v14n2a10","title":"Cadlaws – An English–French Parallel Corpus of Legally Equivalent Documents","year":2021,"lang":"en","type":"article","venue":"Mutatis Mutandis Revista Latinoamericana de Traducción","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Natural language processing; Machine translation; Parallel corpora; Corpus linguistics; Artificial intelligence; Meaning (existential); Linguistics; Translation (biology); Baseline (sea); Psychology; Political science; Law","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004522299,0.0003914683,0.0006931288,0.0001702312,0.0001596124,0.0004560543,0.001408483,0.0001022636,0.0001330341],"category_scores_gemma":[0.00034642,0.0003798189,0.0001829548,0.001224408,0.0002064992,0.0009494302,0.0002781901,0.000371503,0.00000595577],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002849037,"about_ca_system_score_gemma":0.0006203438,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001020454,"about_ca_topic_score_gemma":0.00008504431,"domain_scores_codex":[0.9965086,0.0003997955,0.0007942086,0.0008779177,0.0007605233,0.0006589461],"domain_scores_gemma":[0.997229,0.0001704036,0.0005661698,0.001217417,0.000477543,0.0003395132],"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":[0.0000758907,0.001154837,0.02929483,0.0009555176,0.0004329941,0.003192924,0.02046736,0.0004353828,0.03624326,0.3046386,0.004299349,0.5988091],"study_design_scores_gemma":[0.01595075,0.009184335,0.09758666,0.005149357,0.001613152,0.006019259,0.004306709,0.1409928,0.3201167,0.2698279,0.1126955,0.01655678],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1068425,0.009275606,0.8790284,0.000713748,0.0002947848,0.0005279834,0.00009016252,0.0008885654,0.002338334],"genre_scores_gemma":[0.6109511,0.0001151096,0.3883249,0.0002641141,0.00005919808,0.00002992223,0.00004048214,0.00002849004,0.000186725],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5822523,"threshold_uncertainty_score":0.9998654,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01316222300862668,"score_gpt":0.2824966669699034,"score_spread":0.2693344439612768,"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."}}