{"id":"W4399201246","doi":"10.29173/wclawr112","title":"Innocence Discovery Lab - Harnessing large language models to surface data buried in wrongful conviction case documents","year":2024,"lang":"en","type":"article","venue":"The Wrongful Conviction Law Review","topic":"Digital and Cyber Forensics","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Université Laval; Université du Québec à Montréal","funders":"","keywords":"Innocence; Conviction; Computer science; Political science; Law","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001467722,0.0003075705,0.0003919923,0.00006508608,0.0002232078,0.001180187,0.001314992,0.00006685921,0.00003517154],"category_scores_gemma":[0.00003899837,0.0002245773,0.0000734392,0.001632435,0.00009139867,0.006315179,0.001195718,0.0003269013,0.000339805],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001415156,"about_ca_system_score_gemma":0.000104696,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001259334,"about_ca_topic_score_gemma":0.0007820548,"domain_scores_codex":[0.9973873,0.0002380498,0.0005975708,0.0008606202,0.0004375203,0.0004789511],"domain_scores_gemma":[0.9977741,0.0001222962,0.0001154093,0.001768025,0.00009512735,0.0001251051],"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":[0.00001800722,0.0001617859,0.00003215036,0.001703093,0.0000870693,0.001169167,0.001963273,0.0007285498,0.0001682299,0.9176339,0.007085193,0.06924959],"study_design_scores_gemma":[0.0008131603,0.0001421848,0.00005842608,0.01067488,0.0002010864,0.002144085,0.0004338154,0.1934681,0.0003302796,0.009374595,0.7810985,0.001260962],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08140862,0.186755,0.6720288,0.01772172,0.006280083,0.004427021,0.0007625193,0.001774735,0.02884154],"genre_scores_gemma":[0.9897457,0.003232971,0.0008851186,0.004506172,0.000146569,0.00004693083,0.0001725413,0.00003864646,0.001225286],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9083371,"threshold_uncertainty_score":0.9998567,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03983930074807804,"score_gpt":0.3085035578465523,"score_spread":0.2686642570984742,"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."}}