Reasoning Like an Economist: Post-Training on Economic Problems Induces Strategic Generalization in LLMs

TL;DR

This paper demonstrates that post-training large language models with economic reasoning problems enhances their ability to perform structured multi-agent reasoning and strategic generalization, with implications for real-world economic applications.

Contribution

It introduces Recon, a 7B-parameter LLM fine-tuned on economic problems, showing improved reasoning and strategic generalization in multi-agent scenarios.

Findings

Enhanced economic reasoning performance on benchmarks

Improved multi-agent strategic reasoning capabilities

Post-training with domain-specific data boosts model alignment

Abstract

Directly training Large Language Models (LLMs) for Multi-Agent Systems (MAS) remains challenging due to intricate reward modeling, dynamic agent interactions, and demanding generalization requirements. This paper explores whether post-training techniques, specifically Supervised Fine-Tuning (SFT) and Reinforcement Learning with Verifiable Rewards (RLVR), can effectively $\textit{generalize}$ to multi-agent scenarios. We use economic reasoning as a testbed, leveraging its strong foundations in mathematics and game theory, its demand for structured analytical reasoning, and its relevance to real-world applications such as market design, resource allocation, and policy analysis. We introduce $\textbf{Recon}$ ($\textbf{R}$easoning like an $\textbf{ECON}$omist), a 7B-parameter open-source LLM post-trained on a hand-curated dataset of 2,100 high-quality economic reasoning problems. Comprehensive evaluation on economic reasoning benchmarks and multi-agent games reveals clear improvements in structured reasoning and economic rationality. These results underscore the promise of domain-aligned post-training for enhancing reasoning and agent alignment, shedding light on the roles of SFT and RL in shaping model behavior. Code is available at https://github.com/MasterZhou1/Recon .