There is a newer version of the record available.

Published October 6, 2026 | Version v0.1.1-ae

Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache (Artifact)

  • 1. ROR icon Peking University
  • 2. ROR icon Virginia Tech

Description

This artifact accompanies the paper “Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache,” accepted at the 2027 IEEE Symposium on Security and Privacy.

Files

Cachemir_ae.zip

Files (67.5 kB)

Name Size
md5:1141091adcb77d8382e78f1790c0410d
67.5 kB Preview Download

Additional details

Software

Programming language
Go , Python
Development Status
Active