1 paper
Shangze Li, Chuancheng Shi, Simiao Xie +6
With the widespread deployment of large foundation models (LFMs) in open environments, safety threats are shifting from black-box jailbreaks toward white-box attacks that directly…