Metraipolis › archive › dc_452da40e31d9
Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a Laptop — arXiv:2608.11215
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.11215
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T05:14:43.055Z
- SHA-256 of the archived text
9ad62246d46cf9ade16ba5832626d05773563d438ca95461f4f401f628673b3e- Stable URL of this page
https://metraipolis.com/documento/dc_452da40e31d9
How to cite
Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a Laptop — arXiv:2608.11215 — arXiv (en), https://arxiv.org/abs/2608.11215 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_452da40e31d9 (Metraipolis archive dc_452da40e31d9, sha256 9ad62246d46cf9ade16ba5832626d05773563d438ca95461f4f401f628673b3e)
BibTeX
@misc{dc452da40e31d9,
title = {Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a Laptop — arXiv:2608.11215},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.11215},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_452da40e31d9; sha256 9ad62246d46cf9ade16ba5832626d05773563d438ca95461f4f401f628673b3e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11215. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Simulating societies of many large language model (LLM) agents is expensive, yet the questions asked of such simulations are usually macroscopic: phase behaviour, stylised facts, and scaling with the number of agents $N$, not the cognition of any single agent. We turn a statistical-physics observation into a method: replace each LLM agent by a low-parameter model fitted from a few hundred to a few thousand cheap queries, then run the society at any $N$ on a laptop. Whether this works is decided before the simulation runs, chiefly by what each agent perceives. We introduce an [interaction order x memory] taxonomy that maps perception and memory to an effective theory and a predicted $N$-trend of the surrogate error. We validate it on a faithful reimplementation of the LLM macroeconomy EconAgent and seven further named LLM simulations, with agent decisions cloned from genuine LLM elicitations (primarily DeepSeek) for a few dollars; the predicted error trends hold cell by cell, and the two refuted predictions, both on a strongly saturating response and traced to its curvature, are themselves matched quantitatively by the theory with no free parameters. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.MA). Categorie arXiv: Artificial Intelligence (cs.AI) ; Statistical Mechanics (cond-mat.stat-mech); Computation and Language (cs.CL); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Physics and Society (physics.soc-ph). Pagina del lavoro: https://arxiv.org/abs/2608.11215 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T05:08:22.828Z dalla pagina https://arxiv.org/catchup/cs.MA/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).