inference/research
verifiedhigh · 0.88updated 2026-10-03

QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs (Ashkboos et al.)

paper

Fields

id
sources/qlab-quarot-paper
URL
https://arxiv.org/abs/2404.00456
Publisher
arXiv
Kind
paper
Published
2024-03-30
Accessed
2026-10-03
notes
v2. ETH Zurich SPCL. Randomized Hadamard rotations applied end-to-end including KV cache.

Notes

v2. ETH Zurich SPCL. Randomized Hadamard rotations applied end-to-end including KV cache.

Cross-references

This record references no other records.