[hacker news] KV Cache Compression 900000x Beyond TurboQuant and Per-Vector Shannon Limit]

摘要

KV Cache Compression 900000x Beyond TurboQuant and Per-Vector Shannon Limit]

来源

本文首发于 hacker news。

阅读原文：[KV Cache Compression 900000x Beyond TurboQuant and Per-Vector Shannon Limit]](https://arxiv.org/abs/2604.15356)