{"id":"W4385080367","doi":"10.1109/sp46215.2023.10179396","title":"Private Collaborative Data Cleaning via Non-Equi PSI","year":2023,"lang":"en","type":"article","venue":"","topic":"Cryptography and Data Security","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Bank of Canada","keywords":"Bloom filter; Computer science; Intersection (aeronautics); Exploit; Information leakage; Differential privacy; Set (abstract data type); Data mining; Private information retrieval; Information privacy; Protocol (science); Filter (signal processing); Pseudorandom function family; Data set; Theoretical computer science; Pseudorandom number generator; Computer security; Algorithm; Engineering; Artificial intelligence","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":[],"consensus_categories":[],"category_scores_codex":[0.0004288806,0.0001057516,0.0001160667,0.0001299414,0.0001736905,0.0002309155,0.002251553,0.00004024376,0.00003314322],"category_scores_gemma":[0.00004016264,0.00009340384,0.00002376787,0.002564017,0.00004331802,0.001456795,0.002533032,0.0001038742,0.0005304948],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007185214,"about_ca_system_score_gemma":0.0000461594,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003455084,"about_ca_topic_score_gemma":0.00005619777,"domain_scores_codex":[0.9987725,0.00003511649,0.0001542589,0.0005243575,0.0002314748,0.0002822763],"domain_scores_gemma":[0.9980457,0.00009167573,0.00004809825,0.001669077,0.00005591718,0.0000895356],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001169414,0.00007614224,0.001730092,0.00002329985,0.00005329871,0.00008222769,0.002310323,0.00002924638,0.002403517,0.8073184,0.1449764,0.04098536],"study_design_scores_gemma":[0.0009194693,0.0001881757,0.01327145,0.00004144922,0.00001506675,0.00001225494,0.0005539161,0.632924,0.004689577,0.08618412,0.26039,0.0008104711],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01028766,0.00002520054,0.9805058,0.00108988,0.0004009692,0.000165312,0.0001054729,0.0006815395,0.006738178],"genre_scores_gemma":[0.7636921,0.00009800611,0.2345508,0.0008921672,0.0001560277,0.00001279303,0.0004918837,0.00001561316,0.00009064737],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7534044,"threshold_uncertainty_score":0.6818615,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0315260257843025,"score_gpt":0.2914112050809153,"score_spread":0.2598851792966128,"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."}}