{"id":"W4399151655","doi":"10.48550/arxiv.2405.17878","title":"An Information Theoretic Evaluation Metric For Strong Unlearning","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Statistical Methods and Models","field":"Mathematics","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Instituto de Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México; Institute for Catastrophic Loss Reduction","keywords":"Metric (unit); Computer science; Mathematics; Economics; Operations management","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02631582,0.002818534,0.002352812,0.004995942,0.001478101,0.004275606,0.003023623,0.004126454,0.002068346],"category_scores_gemma":[0.1270899,0.0006454292,0.001461033,0.002970469,0.007064776,0.01054633,0.006315798,0.005632871,0.0006983764],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00417981,"about_ca_system_score_gemma":0.002758744,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008796112,"about_ca_topic_score_gemma":0.0007678143,"domain_scores_codex":[0.9702837,0.0128584,0.002652367,0.003709767,0.009533131,0.0009625863],"domain_scores_gemma":[0.8753197,0.09413081,0.007920269,0.01221547,0.008132273,0.002281507],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00112416,0.0005669387,0.01776754,0.001257997,0.0007535215,0.0002832657,0.0004532329,0.4637612,0.007146465,0.2319766,0.007995343,0.2669137],"study_design_scores_gemma":[0.00004077784,0.0008461239,0.002077874,0.0002404947,0.00009876462,0.0003338742,0.000109143,0.7827163,0.008676542,0.2020957,0.002669944,0.00009444133],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04010387,0.003509698,0.947803,0.001378453,0.0002414061,0.0002424762,0.0007814371,0.0007466028,0.005193095],"genre_scores_gemma":[0.7745913,0.002046341,0.2166552,0.0009943397,0.0005715261,0.0008020995,0.002164465,0.0004075042,0.001767377],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02631582,"threshold_uncertainty_score":0.139173,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2659800695727216,"score_gpt":0.3548664914186651,"score_spread":0.08888642184594348,"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."}}