{"id":"W2955281392","doi":"10.1145/3331184.3331311","title":"On Tradeoffs Between Document Signature Methods for a Legal Due Diligence Corpus","year":2019,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Cisco Systems (Canada)","funders":"","keywords":"Due diligence; Computer science; Hash function; Context (archaeology); Diligence; Signature (topology); Limit (mathematics); Information retrieval; Computer security; Political science; Mathematics; Law; Geography","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":[0.03905644,0.0009373915,0.001478627,0.004443656,0.002266758,0.005585762,0.002342226,0.0025154,0.00257282],"category_scores_gemma":[0.1810595,0.0008855496,0.0009570756,0.004715551,0.001968176,0.008538168,0.003074642,0.001950172,0.001296881],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004038402,"about_ca_system_score_gemma":0.002469831,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004594108,"about_ca_topic_score_gemma":0.007878812,"domain_scores_codex":[0.9742951,0.01283577,0.002213709,0.002163291,0.007276298,0.001215791],"domain_scores_gemma":[0.7270859,0.2270494,0.008993646,0.01817136,0.01582621,0.002873486],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.007423369,0.001069157,0.04359075,0.0008640357,0.0003038076,0.0003852504,0.002019279,0.1887947,0.03214364,0.01644216,0.009314107,0.6976497],"study_design_scores_gemma":[0.0004344542,0.002477489,0.02895727,0.0002980988,0.0002346772,0.001609802,0.003247643,0.8803797,0.04919514,0.02378363,0.009121493,0.000260635],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8308885,0.008737757,0.1411304,0.005183807,0.0002983146,0.000534831,0.0006832677,0.002483586,0.01005963],"genre_scores_gemma":[0.7779223,0.001278152,0.2147762,0.0002817957,0.0001587109,0.0002425818,0.0008386126,0.000601343,0.003900229],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03905644,"threshold_uncertainty_score":0.2065526,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.152014877530545,"score_gpt":0.4960927056453076,"score_spread":0.3440778281147626,"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."}}