Medical large language models are vulnerable to data-poisoning attacks.
Alber DA, Yang Z, Alyakin A, Yang E, Rai S, Valliani AA, Zhang J, Rosenbaum GR, Amend-Thomas AK, Kurland DB, Kremer CM, Eremiev A, Negash B, Wiggan DD, Nakatsuka MA, Sangwon KL, Neifert SN, Khan HA, Save AV, Palla A, Grin EA, Hedman M, Nasir-Moin M, Liu XC, Jiang LY, Mankowski MA, Segev DL, Aphinyanaphongs Y, Riina HA, Golfinos JG, Orringer DA, Kondziolka D, Oermann EK.
Alber DA, et al.
Nat Med. 2025 Jan 8. doi: 10.1038/s41591-024-03445-1. Online ahead of print.
Nat Med. 2025.
PMID: 39779928