{"id":"W3211702131","doi":"10.26615/978-954-452-072-4_139","title":"Exploiting Domain-Specific Knowledge for Judgment Prediction is no Panacea","year":2021,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Research Unit on Children's Psychosocial Maladjustment; Université de Montréal","funders":"Université de Montréal","keywords":"Computer science; Panacea (medicine); Verdict; Artificial intelligence; Task (project management); Scalability; Transformer; Machine learning; Engineering; Law; Political science; Database","routes":{"ca_aff":true,"ca_fund":true,"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":["insufficient_payload"],"category_scores_codex":[0.0006691167,0.00009568808,0.0001255407,0.00003335855,0.0008604829,0.0001326378,0.0001639595,0.0001026764,0.003838297],"category_scores_gemma":[0.0001316909,0.00009868286,0.0001133398,0.0002694081,0.0001813866,0.0002378974,0.00005047946,0.00007835949,0.0009283317],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000219009,"about_ca_system_score_gemma":0.0001819769,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001729792,"about_ca_topic_score_gemma":0.001321162,"domain_scores_codex":[0.9986219,0.0001016473,0.0002891387,0.000329713,0.000270134,0.0003874108],"domain_scores_gemma":[0.9988615,0.0002796742,0.00005697082,0.0001918749,0.0004892199,0.000120776],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00001583266,0.0001592257,0.0006325615,0.00001580861,0.00002444895,0.000003400749,0.03445555,0.000006712173,0.003975038,0.8177162,0.1115693,0.03142586],"study_design_scores_gemma":[0.00005528766,0.00002745859,0.00002399198,0.00002075217,0.00000552557,3.544127e-7,0.03580359,0.0001124294,0.02204161,0.01321888,0.9285781,0.0001119797],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.0463635,0.0005335448,0.1088729,0.004859079,0.003628214,0.0007253325,0.00003303967,0.0003408645,0.8346435],"genre_scores_gemma":[0.8611206,0.0009308803,0.03268402,0.001302215,0.005785674,0.0003320081,0.00003535045,0.0000597574,0.09774947],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.8170088,"threshold_uncertainty_score":0.9998496,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1171803476763433,"score_gpt":0.3732003122322491,"score_spread":0.2560199645559058,"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."}}