{"id":"W4412673258","doi":"10.1145/3731120.3744588","title":"Principles and Guidelines for the Use of LLM Judges","year":2025,"lang":"en","type":"article","venue":"","topic":"Legal Education and Practice Innovations","field":"Social Sciences","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Engineering and Physical Sciences Research Council; Universitas Brawijaya; National Science Foundation","keywords":"Computer science","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":[0.1259661,0.001491356,0.002095619,0.009557955,0.009523786,0.02165887,0.01017394,0.02393229,0.04623748],"category_scores_gemma":[0.2844653,0.003795583,0.002428493,0.01077136,0.01633805,0.01103834,0.00845275,0.02359211,0.05385602],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.007778144,"about_ca_system_score_gemma":0.04462347,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.013444,"about_ca_topic_score_gemma":0.02827931,"domain_scores_codex":[0.8150052,0.09950145,0.02244821,0.005663764,0.05099598,0.006385302],"domain_scores_gemma":[0.7355461,0.1460261,0.009057924,0.03248332,0.0714645,0.005422007],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00008511187,0.0002221425,0.0004421363,0.0009003951,0.00001940522,0.0002174154,0.00810927,0.0004628633,0.0006269352,0.4039956,0.5110189,0.07389978],"study_design_scores_gemma":[0.00008431015,0.00008808352,0.001177393,0.003727124,0.00002879186,0.0004772038,0.00288728,0.001107915,0.0007638464,0.1154334,0.8741024,0.0001222168],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"other","genre_gemma":"methods","genre_scores_codex":[0.003513722,0.008626978,0.2649803,0.1244918,0.01074027,0.01577921,0.005575276,0.003576949,0.5627155],"genre_scores_gemma":[0.04205585,0.008832633,0.5691444,0.02840409,0.00307286,0.04065188,0.004369132,0.002318589,0.3011506],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.1259661,"threshold_uncertainty_score":0.6661803,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.511676247310532,"score_gpt":0.5246279955725135,"score_spread":0.01295174826198153,"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."}}