# Think in Games: Learning to Reason in Games via Reinforcement Learning with Large Language Models

arXiv: https://arxiv.org/abs/2508.21365 · HTML: https://mlsanity.com/paper/2508.21365/

## 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 33 of its 42 references that are in the map.
- First posted on arXiv 29 Aug 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 30 of 33)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning](https://mlsanity.com/paper/2503.09516/) (2025)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Models](https://mlsanity.com/paper/2305.08322/) (2023)
- [Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents](https://mlsanity.com/paper/2302.01560/) (2023)
- [StarCraft II: A New Challenge for Reinforcement Learning](https://mlsanity.com/paper/1708.04782/) (2017)
- [Mastering Complex Control in MOBA Games with Deep Reinforcement Learning](https://mlsanity.com/paper/1912.09729/) (2019)
- [CharacterEval: A Chinese Benchmark for Role-Playing Conversational Agent Evaluation](https://mlsanity.com/paper/2401.01275/) (2024)
- [A Survey on Large Language Model-Based Game Agents](https://mlsanity.com/paper/2404.02039/) (2024)
- [Mind's Eye of LLMs: Visualization-of-Thought Elicits Spatial Reasoning in Large Language Models](https://mlsanity.com/paper/2404.03622/) (2024)
- [Ape210K: A Large-Scale and Template-Rich Dataset of Math Word Problems](https://mlsanity.com/paper/2009.11506/) (2020)
- [CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction](https://mlsanity.com/paper/2502.07316/) (2025)
- [Meta-Reasoner: Dynamic Guidance for Optimized Inference-time Reasoning in Large Language Models](https://mlsanity.com/paper/2502.19918/) (2025)
- [PokeLLMon: A Human-Parity Agent for Pokemon Battles with Large Language Models](https://mlsanity.com/paper/2402.01118/) (2024)
- [Cause and Effect: Can Large Language Models Truly Understand Causality?](https://mlsanity.com/paper/2402.18139/) (2024)
- [MapEval: A Map-Based Evaluation of Geo-Spatial Reasoning in Foundation Models](https://mlsanity.com/paper/2501.00316/) (2024)
- [CharPoet: A Chinese Classical Poetry Generation System Based on Token-free LLM](https://mlsanity.com/paper/2401.03512/) (2024)
- [Creating Pro-Level AI for a Real-Time Fighting Game Using Deep Reinforcement Learning](https://mlsanity.com/paper/1904.03821/) (2019)
- [Enabling Self-Improving Agents to Learn at Test Time With Human-In-The-Loop Guidance](https://mlsanity.com/paper/2507.17131/) (2025)

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

- [Code World Models for General Game Playing](https://mlsanity.com/paper/2510.04542/) (2025)
- [Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Tasks](https://mlsanity.com/paper/2604.20987/) (2026)
- [Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application](https://mlsanity.com/paper/2606.12191/) (2026)
- [Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation](https://mlsanity.com/paper/2601.21464/) (2026)
- [What-If Analysis of Large Language Models: Explore the Game World Using Proactive Thinking](https://mlsanity.com/paper/2509.04791/) (2025)
- [Thinking by Doing: Building Efficient World Model Reasoning in LLMs via Multi-turn Interaction](https://mlsanity.com/paper/2511.23476/) (2025)
- [Stratagem: Learning Transferable Reasoning via Trajectory-Modulated Game Self-Play](https://mlsanity.com/paper/2604.17696/) (2026)

## Abstract (arXiv)

Large language models (LLMs) excel at complex reasoning tasks such as mathematics and coding, yet they frequently struggle with simple interactive tasks that young children perform effortlessly. This discrepancy highlights a critical gap between declarative knowledge (knowing about something) and procedural knowledge (knowing how to do something). Although traditional reinforcement learning (RL) agents can acquire procedural knowledge through environmental interaction, they often operate as black boxes and require substantial training data. In contrast, LLMs possess extensive world knowledge and reasoning capabilities, but are unable to effectively convert this static knowledge into dynamic decision-making in interactive settings. To address this challenge, we propose Think in Games (TiG), a novel framework that empowers LLMs to develop procedural understanding through direct interaction with game environments, while retaining their inherent reasoning and explanatory abilities. Specifically, TiG reformulates RL-based decision-making as a language modeling task: LLMs generate language-guided policies, which are refined iteratively through online reinforcement learning based on environmental feedback. Our experimental results show that TiG successfully bridges the gap between declarative and procedural knowledge, achieving competitive performance with dramatically lower data and computational demands compared to conventional RL methods. Moreover, TiG provides step-by-step natural language explanations for its decisions, greatly improving transparency and interpretability in complex interactive tasks.