---
format: "aidr-story-markdown/v1"
id: "d8e1fb9d8f652d1699ad446e2d1a210d7c910e9ba9d6b8773bd2f8e0156485c6"
canonical_url: "https://aidr.today/d8e1fb9d?lang=en"
title: "To Keep or Not to Keep: Learning KV Cache Retention in Disaggregated LLM Serving Systems"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-22T04:00:00.000Z"
category: "Research"
topics: ["llm"]
source_urls: ["https://arxiv.org/abs/2610.28637"]
summary: "Researchers present KVLearn, a learning-based framework for cost-aware KV-cache retention in disaggregated LLM serving systems. KVLearn uses a Prefix Reuse Predictor, a Cost-Aware Retention Score, and an Adaptive Threshold Controller to improve keep/admit decisions by accounting for recomputation, transfer, and storage costs."
---

# To Keep or Not to Keep: Learning KV Cache Retention in Disaggregated LLM Serving Systems

> [Open the canonical story](<https://aidr.today/d8e1fb9d?lang=en>)

**Published:** 2026-10-22T04:00:00.000Z
**Category:** Research
**Topics:** llm

## Summary

Researchers present KVLearn, a learning\-based framework for cost\-aware KV\-cache retention in disaggregated LLM serving systems\. KVLearn uses a Prefix Reuse Predictor, a Cost\-Aware Retention Score, and an Adaptive Threshold Controller to improve keep/admit decisions by accounting for recomputation, transfer, and storage costs\.

## Sources

- [Story source](<https://arxiv.org/abs/2610.28637>)