{"id":"W3215431778","doi":"10.18653/v1/2021.nllp-1.9","title":"JuriBERT: A Masked-Language Model Adaptation for French Legal Text","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Language model; Domain adaptation; Domain (mathematical analysis); Adaptation (eye); Set (abstract data type); Natural language processing; Focus (optics); Artificial intelligence; Programming language; Psychology; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000757044,0.0002317998,0.0003117876,0.00008483963,0.000466524,0.0005141367,0.0005722429,0.0005169467,0.001182483],"category_scores_gemma":[0.0006188949,0.0002438691,0.0002838505,0.0001625397,0.000268775,0.0002908063,0.0002868419,0.0003873798,0.00006955157],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002456802,"about_ca_system_score_gemma":0.00151285,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.04547765,"about_ca_topic_score_gemma":0.1244687,"domain_scores_codex":[0.9977776,0.0001370791,0.0004366226,0.0006292373,0.0005437293,0.0004757417],"domain_scores_gemma":[0.9985765,0.0002554398,0.0001813589,0.0004341065,0.000404157,0.0001484982],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002561355,0.0002280161,0.00003033531,0.0001380127,0.0001044952,0.00001978805,0.2642507,0.1202951,0.0004675745,0.5219508,0.01149711,0.08099245],"study_design_scores_gemma":[0.00008608556,0.0000343074,0.00001129089,0.0001350585,0.00009512244,7.1535e-7,0.08059209,0.823381,0.002742891,0.06226176,0.02984019,0.0008195089],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03794862,0.0003152727,0.7336671,0.00304593,0.001650376,0.001343948,0.00006975694,0.0002910393,0.2216679],"genre_scores_gemma":[0.8924358,0.00005980648,0.05342616,0.0003768936,0.0007229703,0.0002878705,0.0001197878,0.00003352106,0.05253721],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8544872,"threshold_uncertainty_score":0.9997306,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09036339189723698,"score_gpt":0.3887384578574484,"score_spread":0.2983750659602114,"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."}}