Search papers, labs, and topics across Lattice.
This paper presents a novel method for simulating large societies of language model agents by substituting each agent with a low-parameter model derived from a limited number of queries, enabling efficient simulations on standard laptops. The authors validate their approach through a series of experiments, including a reimplementation of the EconAgent macroeconomic model and seven additional simulations, demonstrating that their predictions regarding surrogate error trends hold true across various scenarios. This method not only reduces computational costs but also provides insights into the interaction dynamics and memory effects within agent societies, revealing the robustness of their theoretical framework against empirical data.
Surprising insights emerge as low-parameter models accurately replicate the behavior of complex LLM-agent societies, challenging the notion that high computational power is essential for meaningful simulations. WHY_IT MATTERS: This approach could democratize access to agent-based modeling, allowing researchers with limited resources to explore complex interactions in LLM societies without sacrificing accuracy.
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.
