{"id":"W4404036747","doi":"10.1109/tifs.2024.3490861","title":"Evaluating Security and Robustness for Split Federated Learning Against Poisoning Attacks","year":2024,"lang":"en","type":"article","venue":"IEEE Transactions on Information Forensics and Security","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of New Brunswick; Queen's University","funders":"","keywords":"Computer science; Robustness (evolution); Computer security","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0007410871,0.0001932523,0.000177587,0.0002351486,0.0007241454,0.001123701,0.0009868889,0.0001676578,0.000003386373],"category_scores_gemma":[0.0005251716,0.000190092,0.00005760713,0.0003890328,0.00009254945,0.00281352,0.0002459585,0.0005427147,0.00000454576],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006905481,"about_ca_system_score_gemma":0.00006477445,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001671026,"about_ca_topic_score_gemma":0.00001486912,"domain_scores_codex":[0.9987355,0.00004458863,0.0003733098,0.0003101099,0.0002584035,0.0002781257],"domain_scores_gemma":[0.9986531,0.0003016119,0.00009471243,0.0007011378,0.0001747934,0.00007470806],"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":[0.00004337333,0.00003078368,0.00001954612,0.0006589741,0.0001023226,0.000003553662,0.004312066,0.01724366,0.0000849182,0.01267445,0.003207429,0.9616189],"study_design_scores_gemma":[0.0003268908,0.0001266432,0.00001173296,0.0001292114,0.00001555744,0.00001834366,0.0002992346,0.9646541,0.001329641,0.03114689,0.00172868,0.0002130334],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1443993,0.00009849056,0.851535,0.00221317,0.0004901811,0.0003160073,0.00006464205,0.000706434,0.0001768323],"genre_scores_gemma":[0.9359264,0.0001646862,0.06366373,0.0001173604,0.000018834,0.00005417382,0.00003808214,0.00001015396,0.000006555509],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9614059,"threshold_uncertainty_score":0.9999132,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03187958657872434,"score_gpt":0.3032457607121363,"score_spread":0.271366174133412,"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."}}