{"id":"W4385653228","doi":"10.14778/3603581.3603585","title":"Autonomously Computable Information Extraction","year":2023,"lang":"en","type":"article","venue":"Proceedings of the VLDB Endowment","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Information extraction; Perspective (graphical); Mechanism (biology); Data mining; Relation (database); Information retrieval; Artificial intelligence","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003585978,0.0000720745,0.00009765203,0.0001475515,0.0001216004,0.0001591556,0.0007093994,0.00002405673,0.000005630097],"category_scores_gemma":[0.00004825445,0.0000515583,0.00006661686,0.0008523987,0.00001855208,0.001337955,0.0003775995,0.00006965925,0.0001332687],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004012356,"about_ca_system_score_gemma":0.00002275906,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005598755,"about_ca_topic_score_gemma":4.753452e-7,"domain_scores_codex":[0.9992123,0.000003205348,0.0002170793,0.0001231146,0.0002829995,0.000161292],"domain_scores_gemma":[0.9994956,0.00002048164,0.000199124,0.0001420602,0.0001108833,0.00003180914],"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":[0.000024766,0.0002862445,0.01332742,0.0004480376,0.0003277726,0.000001502454,0.01067733,0.008057961,0.07466847,0.2894354,0.2375818,0.3651632],"study_design_scores_gemma":[0.0008345249,0.0001420553,0.02117872,0.000167215,0.00007103456,0.00003063203,0.001127974,0.6878336,0.1598304,0.0111902,0.1171496,0.0004440111],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8391898,0.0000694075,0.06641932,0.02073203,0.002290572,0.001046871,0.00003457318,0.002219438,0.06799798],"genre_scores_gemma":[0.9933143,0.00001570547,0.006013122,0.0001218082,0.00003174732,0.00001677462,0.000004449399,0.000003154154,0.0004789555],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6797757,"threshold_uncertainty_score":0.2102486,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01020545676619336,"score_gpt":0.2254807868001845,"score_spread":0.2152753300339911,"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."}}