{"id":"W4401719912","doi":"10.1109/re59067.2024.00051","title":"Rethinking Legal Compliance Automation: Opportunities with Large Language Models","year":2024,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"Mitacs","keywords":"Compliance (psychology); Automation; Computer science; Engineering ethics; Engineering; Psychology","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":[],"category_scores_codex":[0.0006460987,0.00008326902,0.00009052983,0.00004942754,0.0004852057,0.0004250626,0.0002039597,0.00006225747,0.001509983],"category_scores_gemma":[0.00002799481,0.00006763743,0.00003341732,0.0001787077,0.0002789153,0.001057412,0.00003080742,0.0001376852,0.0001913827],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007534515,"about_ca_system_score_gemma":0.0002544913,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002630442,"about_ca_topic_score_gemma":0.01166921,"domain_scores_codex":[0.9988943,0.00007780314,0.0001486577,0.000186871,0.0004046721,0.0002876389],"domain_scores_gemma":[0.9995562,0.0001273523,0.00002485563,0.0001363521,0.00007795711,0.00007728237],"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":[0.000002388837,0.000008700234,0.000009265706,0.00001456258,0.000008685536,0.00007322502,0.06119428,0.0001337082,0.00001293652,0.9314963,0.001976321,0.005069667],"study_design_scores_gemma":[0.00005236087,0.00006135719,0.00001027618,0.0004915511,0.00002761515,0.00001368802,0.1534982,0.2477443,0.001004842,0.1937027,0.4028932,0.0005000262],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.01992228,0.0004577955,0.09826985,0.007856386,0.0004939294,0.0002024526,0.00001362586,0.001302558,0.8714811],"genre_scores_gemma":[0.9611847,0.00003381525,0.002653512,0.0005141015,0.000268964,0.00001165608,0.000005821634,0.00001329873,0.03531415],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9412624,"threshold_uncertainty_score":0.9994028,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2263741333273775,"score_gpt":0.3931620310209396,"score_spread":0.1667878976935621,"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."}}