@article{revisitingblockbasedquantisationwhatis, title = {Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference?}, author = {Cheng Zhang and Jianyi Cheng and Ilia Shumailov and George A. Constantinides and Yiren Zhao}, year = {2023}, eprint = {2310.05079}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2310.05079v2}, }