{"id":"W6903066260","doi":"10.1016/j.neucom.2025.130979","title":"MKE-PLLM: A benchmark for multilingual knowledge editing on pretrained large language model","year":2025,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Major Science and Technology Projects in Yunnan Province; Applied Basic Research Foundation of Yunnan Province; Yunnan Provincial Science and Technology Department; National Natural Science Foundation of China","keywords":"Benchmark (surveying); Usability; Robustness (evolution); Focus (optics); Language model; Baseline (sea)","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":[0.005858644,0.004140384,0.002349084,0.004736313,0.002000519,0.00413227,0.007636585,0.004624597,0.02588649],"category_scores_gemma":[0.03149627,0.001451291,0.002466347,0.004336527,0.001050953,0.00827892,0.005370782,0.004486872,0.0158851],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00212524,"about_ca_system_score_gemma":0.004012132,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02144191,"about_ca_topic_score_gemma":0.0322669,"domain_scores_codex":[0.9937339,0.001797995,0.0007758848,0.001910399,0.001162485,0.0006192496],"domain_scores_gemma":[0.9837005,0.008326531,0.0003561938,0.004095045,0.00277825,0.0007435539],"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":[0.002447477,0.001711979,0.002952356,0.002788338,0.001320777,0.001027662,0.0003782278,0.07466412,0.007916692,0.004071732,0.2401072,0.6606134],"study_design_scores_gemma":[0.001268402,0.000961373,0.004036514,0.0003744001,0.0005033383,0.00116141,0.0009365422,0.8584063,0.04182658,0.01626198,0.07399677,0.0002663544],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"software","genre_gemma":"empirical","genre_scores_codex":[0.2305366,0.01465769,0.239719,0.002801784,0.004436209,0.001817098,0.100146,0.3562827,0.049603],"genre_scores_gemma":[0.3010024,0.002098854,0.3916472,0.001144526,0.0003298498,0.001190083,0.2737671,0.01142132,0.01739868],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02588649,"threshold_uncertainty_score":0.08659887,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01384809728640121,"score_gpt":0.3358910638878189,"score_spread":0.3220429666014177,"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."}}