Game Theory Meets Large Language Models: A Systematic Survey with Taxonomy and New Frontiers

Game theory is a foundational framework for analyzing strategic interactions, and its intersection with large language models (LLMs) is a rapidly growing field. However, existing surveys mainly focus narrowly on using game theory to evaluate LLM behavior. This paper provides the first comprehensive survey of the bidirectional relationship between Game Theory and LLMs. We propose a novel taxonomy that categorizes the research in this intersection into four distinct perspectives: (1) evaluating LLMs in game-based scenarios; (2) improving LLMs using game-theoretic concepts for better interpretability and alignment; (3) modeling the competitive landscape of LLM development and its societal impact; and (4) leveraging LLMs to advance game models and to solve corresponding game theory problems. Furthermore, we identify key challenges and outline future research directions. By systematically investigating this interdisciplinary landscape, our survey highlights the mutual influence of game theory and LLMs, fostering progress at the intersection of these fields.

• Publication year

  2025

• Venue

  Unknown venue

• Publication date

  2025-02-13

• Fields of study

  Computer Science

• Identifiers
  arXiv 2502.09053
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• An Automated Reinforcement Learning Reward Design Framework with Large Language Model for Cooperative Platoon Coordination

  2025cited by this paper
• FAIRGAME: a Framework for AI Agents Bias Recognition using Game Theory

  2025cited by this paper
• Incentivizing High-Quality Human Annotations with Golden Questions

  2025cited by this paper
• Heterogeneous Data Game: Characterizing the Model Competition Across Multiple Data Sources

  2025cited by this paper
• Clone-Robust AI Alignment

  2025cited by this paper
• Negotiative Alignment: Embracing Disagreement to Achieve Fairer Outcomes - Insights from Urban Studies

  2025cited by this paper
• Extragradient Preference Optimization (EGPO): Beyond Last-Iterate Convergence for Nash Learning from Human Feedback

  2025cited by this paper
• Accelerating Nash Learning from Human Feedback via Mirror Prox

  2025cited by this paper
• Reward-Guided Prompt Evolving in Reinforcement Learning for LLMs

  2025cited by this paper
• Incentivizing Truthful Language Models via Peer Elicitation Games

  2025cited by this paper
• Stackelberg Self-Annotation: A Robust Approach to Data-Efficient LLM Alignment

  2025cited by this paper
• PokerBench: Training Large Language Models to become Professional Poker Players

  2025cited by this paper
• Understanding and Optimizing Agentic Workflows via Shapley value

  2025cited by this paper
• Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium

  2025cited by this paper
• The Economics of Large Language Models: Token Allocation, Fine-Tuning, and Optimal Pricing

  2025cited by this paper
• Strategyproof Reinforcement Learning from Human Feedback

  2025cited by this paper
• Too Much Information? Investigating Information Disclosure in Auction Systems with LLM Simulations

  2025cited by this paper
• Advertising in AI systems: Society must be vigilant

  2025cited by this paper
• Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Models

  2025cited by this paper
• Game of Thoughts: Iterative Reasoning in Game-Theoretic Domains with Large Language Models

  2025cited by this paper
• Fair Document Valuation in LLM Summaries via Shapley Values

  2025cited by this paper
• SCAR: Shapley Credit Assignment for More Efficient RLHF

  2025cited by this paper
• The Backfiring Effect of Weak AI Safety Regulation

  2025cited by this paper
• Flattening Supply Chains: When do Technology Improvements lead to Disintermediation?

  2025cited by this paper
• Verbalized Bayesian Persuasion

  2025cited by this paper
• From Natural Language to Extensive-Form Game Representations

  2025cited by this paper
• Accelerated Preference Elicitation with LLM-Based Proxies

  2025cited by this paper
• Selective Response Strategies for GenAI

  2025cited by this paper
• Human Decision-making is Susceptible to AI-driven Manipulation

  2025cited by this paper
• An Interpretable Automated Mechanism Design Framework with Large Language Models

  2025cited by this paper
• Distributive Fairness in Large Language Models: Evaluating Alignment with Human Values

  2025cited by this paper
• How Humans Help LLMs: Assessing and Incentivizing Human Preference Annotators

  2025cited by this paper
• Direct Alignment with Heterogeneous Preferences

  2025cited by this paper
• Playing games with Large language models: Randomness and strategy

  2025cited by this paper
• Empowering LLMs in Decision Games through Algorithmic Data Synthesis

  2025cited by this paper
• Navigating the Deployment Dilemma and Innovation Paradox: Open-Source versus Closed-source Models

  2025cited by this paper
• Efficient Shapley Value-based Non-Uniform Pruning of Large Language Models

  2025cited by this paper
• Strategic Content Creation with Age of GenAI: To Share or Not to Share?

  2025cited by this paper
• Generative Social Choice: The Next Generation

  2025cited by this paper
• Beyond RLHF and NLHF: Population-Proportional Alignment under an Axiomatic Framework

  2025cited by this paper
• Theoretical Tensions in RLHF: Reconciling Empirical Success with Inconsistencies in Social Choice Theory

  2025cited by this paper
• Will Systems of LLM Agents Cooperate: An Investigation into a Social Dilemma

  2025cited by this paper
• Beyond Nash Equilibrium: Bounded Rationality of LLMs and humans in Strategic Decision-making

  2025cited by this paper
• From Debate to Equilibrium: Belief-Driven Multi-Agent LLM Reasoning via Bayesian Nash Equilibrium

  2025cited by this paper
• Distortion of AI Alignment: Does Preference Optimization Optimize for Preferences?

  2025cited by this paper
• Fundamental Limits of Game-Theoretic LLM Alignment: Smith Consistency and Preference Matching

  2025cited by this paper
• SPC: Evolving Self-Play Critic via Adversarial Games for LLM Reasoning

  2025cited by this paper
• Steering Language Models with Game-Theoretic Solvers

  2024cited by this paper
• PokeLLMon: A Human-Parity Agent for Pokemon Battles with Large Language Models

  2024cited by this paper
• Simulating Human Strategic Behavior: Comparing Single and Multi-agent LLMs

  2024cited by this paper
• Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents

  2024cited by this paper
• Economics Arena for Large Language Models

  2024cited by this paper
• Ad Auctions for LLMs via Retrieval Augmented Generation

  2024cited by this paper
• MaxMin-RLHF: Towards Equitable Alignment of Large Language Models with Diverse Human Preferences

  2024cited by this paper
• Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models

  2024cited by this paper
• A Minimaximalist Approach to Reinforcement Learning from Human Feedback

  2024cited by this paper
• Evaluating Language Model Agency through Negotiations

  2024cited by this paper
• PokerGPT: An End-to-End Lightweight Solver for Multi-Player Texas Hold'em via Large Language Model

  2024influential reference
• VLN-Game: Vision-Language Equilibrium Search for Zero-Shot Semantic Navigation

  2024cited by this paper
• K-Level Reasoning: Establishing Higher Order Beliefs in Large Language Models for Strategic Reasoning

  2024cited by this paper
• Rethinking Strategic Mechanism Design In The Age Of Large Language Models: New Directions For Communication Systems

  2024cited by this paper
• How Well Can LLMs Negotiate? NegotiationArena Platform and Analysis

  2024cited by this paper
• Rethinking Bradley-Terry Models in Preference-Based Reward Modeling: Foundations, Theory, and Alternatives

  2024cited by this paper
• Language Alignment via Nash-learning and Adaptive feedback

  2024cited by this paper
• Game-theoretic LLM: Agent Workflow for Negotiation Games

  2024cited by this paper
• A Survey on Large Language Model-Based Social Agents in Game-Theoretic Scenarios

  2024cited by this paper
• Learning from Response not Preference: A Stackelberg Approach for LLM Detoxification using Non-parallel Data

  2024cited by this paper
• Magnetic Preference Optimization: Achieving Last-iterate Convergence for Language Models Alignment

  2024cited by this paper
• Representative Social Choice: From Learning Theory to AI Alignment

  2024cited by this paper
• Evaluating Large Language Models with Grid-Based Game Competitions: An Extensible LLM Benchmark and Leaderboard

  2024cited by this paper
• Generative AI as Economic Agents

  2024cited by this paper
• Incentivizing Quality Text Generation via Statistical Contracts

  2024cited by this paper
• Natural Language Mechanisms via Self-Resolution with Foundation Models

  2024cited by this paper
• Pricing and Competition for Generative AI

  2024cited by this paper
• DeepSeek-V3 Technical Report

  2024cited by this paper
• Policy Aggregation

  2024cited by this paper
• Proportional Representation for Artificial Intelligence

  2024cited by this paper
• TMGBench: A Systematic Game Benchmark for Evaluating Strategic Reasoning Abilities of LLMs

  2024cited by this paper
• Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning

  2024cited by this paper
• Mechanism Design for Large Language Models

  2024cited by this paper
• Coevolving with the Other You: Fine-Tuning LLM with Sequential Cooperative Multi-Agent Reinforcement Learning

  2024cited by this paper
• Linguistically Grounded Analysis of Language Models using Shapley Head Values

  2024cited by this paper
• Self-Play Preference Optimization for Language Model Alignment

  2024cited by this paper
• RLHF from Heterogeneous Feedback via Personalization and Preference Aggregation

  2024cited by this paper
• Investigating Layer Importance in Large Language Models

  2024cited by this paper
• Mapping Social Choice Theory to RLHF

  2024cited by this paper
• SHED: Shapley-Based Automated Dataset Refinement for Instruction Fine-Tuning

  2024cited by this paper
• Algorithmic Decision-Making under Agents with Persistent Improvement

  2024cited by this paper
• Truthful Aggregation of LLMs with an Application to Online Advertising

  2024cited by this paper
• Axioms for AI Alignment from Human Feedback

  2024cited by this paper
• Eliciting Informative Text Evaluations with Large Language Models

  2024cited by this paper
• Mechanism Design for LLM Fine-tuning with Multiple Reward Models

  2024cited by this paper
• STRIDE: A Tool-Assisted LLM Agent Framework for Strategic and Interactive Decision-Making

  2024cited by this paper
• Learning to Discuss Strategically: A Case Study on One Night Ultimate Werewolf

  2024cited by this paper
• Self-playing Adversarial Language Game Enhances LLM Reasoning

  2024cited by this paper
• Contextual Bilevel Reinforcement Learning for Incentive Alignment

  2024cited by this paper
• How to Strategize Human Content Creation in the Era of GenAI?

  2024cited by this paper
• PAL: Pluralistic Alignment Framework for Learning from Heterogeneous Preferences

  2024cited by this paper
• ReTA: Recursively Thinking Ahead to Improve the Strategic Reasoning of Large Language Models

  2024cited by this paper
• Autoformalization of Game Descriptions using Large Language Models

  2024cited by this paper

• When Personas Override Payoffs: Role Identity Bias in Multi-Agent LLM Decision-Making

  2026influential citation
• Discovering Differences in Strategic Behavior Between Humans and LLMs

  2026cites this paper
• AIDG: Evaluating Asymmetry Between Information Extraction and Containment in Multi-Turn Dialogue

  2026cites this paper
• Communication Enables Cooperation in LLM Agents: A Comparison with Curriculum-Based Approaches

  2025cites this paper
• GTAlign: Game-Theoretic Alignment of LLM Assistants for Mutual Welfare

  2025cites this paper
• Strategic Learning Under Linguistic and Contextual Constraints: A Theoretical Framework for LLM-Based Multi-Agent Coordination

  2025cites this paper