{"id":"W4385571145","doi":"10.18653/v1/2023.acl-long.865","title":"LeXFiles and LegalLAMA: Facilitating English Multinational Legal Language Model Development","year":2023,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fonds de recherche du Québec – Nature et technologies; Innovationsfonden","keywords":"Multinational corporation; Upstream (networking); Benchmark (surveying); Computer science; Downstream (manufacturing); Best practice; Natural language processing; Work (physics); Knowledge management; Artificial intelligence; Engineering; Political science; Operations management; Law","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["open_science"],"consensus_categories":[],"category_scores_codex":[0.005826701,0.001274113,0.0005654818,0.001437208,0.0006599394,0.002148102,0.003176749,0.001295041,0.01548059],"category_scores_gemma":[0.03336447,0.0007457674,0.0007956204,0.001098481,0.0007969061,0.006813524,0.003528602,0.003191066,0.00633906],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001686596,"about_ca_system_score_gemma":0.002524406,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02186923,"about_ca_topic_score_gemma":0.03334502,"domain_scores_codex":[0.9971713,0.001356913,0.0002005088,0.0006295155,0.0004871754,0.0001546011],"domain_scores_gemma":[0.9837492,0.01063299,0.0003791148,0.003116261,0.001703949,0.0004185147],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001076681,0.001689576,0.02365947,0.001522735,0.000354157,0.0008906933,0.003023647,0.1108214,0.01647819,0.0139379,0.1872223,0.6393232],"study_design_scores_gemma":[0.0003013975,0.000529913,0.009581777,0.0003401523,0.00009656923,0.000518275,0.001966809,0.8347487,0.03618428,0.01098132,0.1045817,0.0001690267],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"dataset","genre_scores_codex":[0.5167446,0.001745508,0.2054652,0.004497848,0.0008771823,0.00127434,0.03893145,0.1923929,0.03807082],"genre_scores_gemma":[0.6019179,0.0005890641,0.2607052,0.000919559,0.00007512816,0.001121614,0.1165643,0.007881297,0.01022584],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.9968233,"threshold_uncertainty_score":0.05178767,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01405885961727889,"score_gpt":0.2717166111702214,"score_spread":0.2576577515529425,"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."}}