{"id":"W4226246254","doi":"10.1609/aaai.v36i11.21438","title":"Interpretable Low-Resource Legal Decision Making","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"Government of Canada","keywords":"Interpretability; Deep learning; Artificial intelligence; Computer science; Machine learning; Task (project management); Resource (disambiguation); Confusion; Trademark; Data science; Engineering","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":["metaepi_narrow","sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002335238,0.0002985609,0.0003763578,0.0002373535,0.002489927,0.000408407,0.003043845,0.000122428,0.004662912],"category_scores_gemma":[0.001938031,0.0002671871,0.0002737095,0.001417379,0.001240938,0.0005242605,0.001034645,0.0008513432,0.000169035],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003788496,"about_ca_system_score_gemma":0.0003565798,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005746634,"about_ca_topic_score_gemma":0.0003052088,"domain_scores_codex":[0.9957491,0.0001236586,0.0009424267,0.0006715244,0.001734756,0.0007785147],"domain_scores_gemma":[0.9977612,0.0005004007,0.0006103594,0.000369161,0.000606802,0.0001520545],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0003293042,0.000235195,0.0003173706,0.00001312437,0.00001623825,0.00000178917,0.008874246,0.0008543114,0.004768332,0.8955095,0.001672164,0.0874084],"study_design_scores_gemma":[0.00004740862,0.0005761441,0.00007655571,0.0008184582,0.00006719517,0.00001351262,0.08423933,0.01924628,0.1874824,0.6352114,0.07116959,0.001051733],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6188173,0.00006375254,0.004193867,0.007049133,0.002512356,0.001435255,0.00003655208,0.0002958126,0.365596],"genre_scores_gemma":[0.996376,0.00001905193,0.0006023357,0.0006771981,0.0002385016,0.00009728164,7.341855e-7,0.00003414618,0.001954796],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3775586,"threshold_uncertainty_score":0.999978,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06607011868209936,"score_gpt":0.3496311093982654,"score_spread":0.283560990716166,"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."}}