# Approximating Human Strategic Reasoning with LLM-Enhanced Recursive Reasoners Leveraging Multi-agent Hypergames

arXiv: https://arxiv.org/abs/2502.07443 · HTML: https://mlsanity.com/paper/2502.07443/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 12 of its 49 references that are in the map.
- First posted on arXiv 11 Feb 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)
- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)

## Builds on (most cited first, 12 of 12)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [GTBench: Uncovering the Strategic Reasoning Limitations of LLMs via Game-Theoretic Evaluations](https://mlsanity.com/paper/2402.12348/) (2024)
- [Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/1901.09207/) (2019)
- [Towards Better Chain-of-Thought Prompting Strategies: A Survey](https://mlsanity.com/paper/2310.04959/) (2023)
- [ALYMPICS: LLM Agents Meet Game Theory -- Exploring Strategic Decision-Making with AI Agents](https://mlsanity.com/paper/2311.03220/) (2023)
- [LLM-based Human Simulations Have Not Yet Been Reliable](https://mlsanity.com/paper/2501.08579/) (2025)
- [K-Level Reasoning: Establishing Higher Order Beliefs in Large Language Models for Strategic Reasoning](https://mlsanity.com/paper/2402.01521/) (2024)
- [Economics Arena for Large Language Models](https://mlsanity.com/paper/2401.01735/) (2024)
- [Towards Logically Sound Natural Language Reasoning with Logic-Enhanced Language Model Agents](https://mlsanity.com/paper/2408.16081/) (2024)
- [Generative Agents for Multi-Agent Autoformalization of Interaction Scenarios](https://mlsanity.com/paper/2412.08805/) (2024)

## Built on by (most cited first, 7 of 7)

- [A Survey of Frontiers in LLM Reasoning: Inference Scaling, Learning to Reason, and Agentic Systems](https://mlsanity.com/paper/2504.09037/) (2025)
- [The Influence of Human-inspired Agentic Sophistication in LLM-driven Strategic Reasoners](https://mlsanity.com/paper/2505.09396/) (2025)
- [Matching Markets Meet LLMs: Algorithmic Reasoning with Ranked Preferences](https://mlsanity.com/paper/2506.04478/) (2025)
- [A Survey on Hypergame Theory: Modelling Misaligned Perceptions and Nested Beliefs for Multi-Agent Systems](https://mlsanity.com/paper/2507.19593/) (2025)
- [Interview-Informed Generative Agents for Product Discovery: A Validation Study](https://mlsanity.com/paper/2603.29890/) (2026)
- [Hypergame Rationalisability: Solving Agent Misalignment In Strategic Play](https://mlsanity.com/paper/2512.11942/) (2025)
- [Clueing up LLMs with Tool-Augmented Deductive Reasoning](https://mlsanity.com/paper/2609.18736/) (2026)

## Abstract (arXiv)

LLM-driven multi-agent-based simulations have been gaining traction with applications in game-theoretic and social simulations. While most implementations seek to exploit or evaluate LLM-agentic reasoning, they often do so with a weak notion of agency and simplified architectures. We implement a role-based multi-agent strategic interaction framework tailored to sophisticated recursive reasoners, providing the means for systematic in-depth development and evaluation of strategic reasoning. Our game environment is governed by the umpire responsible for facilitating games, from matchmaking through move validation to environment management. Players incorporate state-of-the-art LLMs in their decision mechanism, relying on a formal hypergame-based model of hierarchical beliefs. We use one-shot, 2-player beauty contests to evaluate the recursive reasoning capabilities of the latest LLMs, providing a comparison to an established baseline model from economics and data from human experiments. Furthermore, we introduce the foundations of an alternative semantic measure of reasoning to the k-level theory. Our experiments show that artificial reasoners can outperform the baseline model in terms of both approximating human behaviour and reaching the optimal solution.