{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0006829731,0.0002333646,0.0002246306,0.0002316951,0.0005413794,0.0001601348,0.0009757519,0.00008791759,0.000001945133],"category_scores_gemma":[0.0008430019,0.0002452501,0.0001276513,0.0005299944,0.000020988,0.000116805,0.0004644621,0.000433809,0.0000192436],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005457082,"about_ca_system_score_gemma":0.0001170144,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001772329,"about_ca_topic_score_gemma":0.000007714104,"domain_scores_codex":[0.9980505,0.000110846,0.0003545287,0.0007692425,0.0001703105,0.0005446507],"domain_scores_gemma":[0.9979029,0.001114234,0.000141231,0.0006315768,0.0001234567,0.00008660804],"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":[0.00002868883,0.0007339072,0.001478142,0.0003217992,0.00004938244,0.00001034605,0.01162386,0.2084714,0.01264344,0.1941134,0.002448276,0.5680774],"study_design_scores_gemma":[0.0007271238,0.00004464014,0.001461324,0.00008472874,0.000007716395,0.000001938287,0.00004829634,0.9935567,0.001532737,0.0007557743,0.001603424,0.0001756001],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1810026,0.00003246684,0.8079965,0.0007005068,0.0003685781,0.0006317678,0.00001271609,0.0006335404,0.008621247],"genre_scores_gemma":[0.882193,2.917061e-7,0.116142,0.0008823007,0.0002988785,0.00008537846,0.00001236789,0.00002229769,0.0003634358],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7850853,"threshold_uncertainty_score":1,"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."}}