Here are three papers describing different side-channel attacks against LLMs. “Remote Timing Attacks on Efficient Language Model Inference“: Abstract: Scaling up language models has significantly ...

A technical paper titled “A New Security Threat in MCUs – SoC-wide timing side channels and how to find them” was published by researchers at University of Kaiserslautern-Landau and Stanford ...

Hertzbleed is a new side-channel attack that turns a power side channel into a timing side channel. That timing side channel may be exploitable even if the algorithm runs in a constant number of clock ...