2026-07-02 04:00 UTCOriginal source2 min readUpdated: 2026-07-02 07:59 UTC

From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents

A study on Lewis signaling games with LLM agents shows that memory architecture, specifically a persistent private notebook, significantly improves coordination over stateless agents, and channel capacity alone cannot predict success.

SourcearXiv AIAuthor: Yashar Talebirad, Eden Redman, Ali Parsaee, Osmar R. Zaiane

-->

[Submitted on 30 Jun 2026]

Title:From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents

View a PDF of the paper titled From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents, by Yashar Talebirad and 3 other authors

View PDF HTML (experimental)

Abstract:How do two agents invent a shared language from scratch? In a Lewis signaling game, a sender and receiver must coordinate on a code using only their interaction history. We study five memory architectures across varying channel configurations with LLM agents and find that memory architecture matters more than channel capacity. Agents with a persistent private notebook benefit from surplus channel capacity and avoid the high-capacity collapse seen in stateless agents, achieving the most reliable coordination ($0.867 \pm 0.023$ at capacity = 25). Stateless agents peak at moderate capacity and then degrade as the vocabulary grows beyond what a rolling context window can track The notebook externalizes learned conventions, freeing agents from having to re-derive codes each round. An information bottleneck-inspired argument predicts an optimal capacity equal to the number of objects. Instead, the bottleneck (capacity = 8) proves to be a fragility point, and surplus capacity is generally better. We show that channel capacity alone cannot predict coordination; memory architecture determines whether agents turn interaction history into stable conventions, and both dimensions are needed to understand how signals become language.

Subjects:

Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Theory (cs.IT); Multiagent Systems (cs.MA)

Cite as: arXiv:2607.00233 [cs.AI]

(or arXiv:2607.00233v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2607.00233

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yashar Talebirad [view email] [v1] Tue, 30 Jun 2026 22:20:41 UTC (1,133 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents, by Yashar Talebirad and 3 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.AI

new | recent | 2026-07

Change to browse by:

cs cs.CL cs.IT cs.MA math math.IT

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)