{"id":"W3186932802","doi":"10.1145/3462757.3466144","title":"Labels distribution matters in performance achieved in legal judgment prediction tasks","year":2021,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Lawsuit; Computer science; Task (project management); Artificial intelligence; Transformer; Representation (politics); Natural language processing; Intersection (aeronautics); Machine learning; Law; Political science; Engineering","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":[0.006709018,0.002087933,0.001395853,0.00166803,0.001477458,0.00409517,0.002260434,0.004497012,0.01221613],"category_scores_gemma":[0.04180127,0.0006472434,0.0007682666,0.001759634,0.001334041,0.009166032,0.002779702,0.004902423,0.02179359],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001552556,"about_ca_system_score_gemma":0.001562122,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006932592,"about_ca_topic_score_gemma":0.00890047,"domain_scores_codex":[0.9940784,0.001593635,0.0003464559,0.002247076,0.001187905,0.0005464709],"domain_scores_gemma":[0.9792739,0.01402843,0.0006517195,0.003263771,0.001815849,0.000966296],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.004825958,0.002236288,0.02871425,0.000976634,0.0003114275,0.0004786203,0.000624038,0.04451933,0.01671509,0.005703813,0.1119646,0.78293],"study_design_scores_gemma":[0.0009087675,0.002604525,0.04226273,0.0006702151,0.0004829296,0.00116511,0.002142595,0.7078134,0.0465312,0.1325803,0.06258777,0.0002505532],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5611967,0.02792093,0.1972546,0.02754077,0.004970265,0.0007204708,0.01901915,0.03798194,0.1233952],"genre_scores_gemma":[0.9100673,0.002613796,0.03805642,0.002354245,0.0009754595,0.0001999649,0.02461681,0.003383997,0.01773201],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01221613,"threshold_uncertainty_score":0.04086697,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03060566189741133,"score_gpt":0.3112505753430578,"score_spread":0.2806449134456465,"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."}}