---
title: "Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents — Stuff That Spins"
description: "arXiv:2607.05690v1 Announce Type: new Abstract: Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store qu…"
	canonical: "https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents"
html: "https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents"
json: "https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents.json"
markdown: "https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents.md"
keywords: ["narrative intelligence", "SpinGraph", "AI recall"]
date: "2026-07-08T04:00:00+00:00"
modified: "2026-07-08T06:02:32.447492+00:00"
json_ld: |
  {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://stuffthatspins.com/#organization","name":"Stuff That Spins","url":"https://stuffthatspins.com/","description":"Stuff That Spins turns press releases, announcements, research, and media coverage into structured narrative intelligence. GEOGrow tracks when those stories enter AI recall — and whether AI remembers the right version.","logo":{"@type":"ImageObject","url":"https://stuffthatspins.com/images/logo.png"},"sameAs":[]},{"@type":"NewsArticle","@id":"https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents#article","headline":"Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents","description":"arXiv:2607.05690v1 Announce Type: new Abstract: Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store qu…","datePublished":"2026-07-08T04:00:00+00:00","dateModified":"2026-07-08T06:02:32.447492+00:00","url":"https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents","mainEntityOfPage":{"@type":"WebPage","@id":"https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents"},"isAccessibleForFree":true,"inLanguage":"en-US","articleSection":"research","author":{"@type":"Organization","name":"arXiv Artificial Intelligence","url":"https://export.arxiv.org/rss/cs.AI"},"publisher":{"@id":"https://stuffthatspins.com/#organization"},"citation":"https://arxiv.org/abs/2607.05690","about":[],"mentions":[{"@type":"Organization","name":"arXiv Artificial Intelligence"}]},{"@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Stuff That Spins","item":"https://stuffthatspins.com/"},{"@type":"ListItem","position":2,"name":"Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents","item":"https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents"}]}]}
---

# Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents

**Source:** Unknown  
**Published:** July 8, 2026  
**Original:** https://arxiv.org/abs/2607.05690  

## On this page

- [Overview](#overview)

<a id="overview"></a>

## Overview

arXiv:2607.05690v1 Announce Type: new Abstract: Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store queried at most once per turn. We study the regime where memory moves inside the loop, read and written on every step. The obstacle has always been latency: networked stores answer in tens to hundreds of milliseconds, and in-loop retrieval can inflate end-to-end latency by up to 83x when retrieval is expensive. Prior work manages that cost r

---
*HTML version: https://stuffthatspins.com/spin/memory-in-the-loop-in-process-retrieval-as-extendedworking-memory-for-language-agents*