Durable Quantization Conditioned Misalignment Attack on Large Language Models.

Peiran Dong, Haowei Li, Song Guo

Browse the full ICLR paper archive.