Through PlayTrain, an RL framework that combines the abilities of large language models to robustly generate JavaScript games from a minimal human prompt, and an efficient pipeline that can run any JS game in a standard'gym'environment, this work reimagine RL VGE development.
Abstract
While many video-game environments (VGEs) have played crucial roles in advancing reinforcement learning (RL), developing novel VGEs or modifying existing ones to support new features, has been a laborious process requiring extensive hand-coding. Here we present PlayTrain, an RL framework that combines the abilities of large language models (LLMs) to robustly generate JavaScript (JS) games from a minimal human prompt, and an efficient pipeline that can run any JS game in a standard'gym'environment. Not only are recent LLMs particularly good at writing JS code, but the JS format also allows users to easily play generated VGEs, while PlayTrain enables us to train RL agents on the exact same games. We demonstrate multiple use cases of PlayTrain, including cloning well-known Atari and ProcGen games in simple JS, where PlayTrain trains pixel-based agents end-to-end at over 1M agent-decisions per second on a single GPU node; and creating modified versions thereof (e.g., that support novel test sets, procedural generation logics, or game dynamics). Through PlayTrain, we reimagine RL VGE development: all we need is a single JS file, generated and modified through an LLM. We discuss promising future RL research directions that PlayTrain unlocks.
A runtime strategy-selection framework in which a large language model (LLM) guides a trained RL policy without modifying its underlying behavior is investigated, demonstrating both the potential and limitations of LLM-guided runtime strategy selection for adaptive multi-agent game AI.
River, a simple training recipe that improves reward quality by filtering low-quality environments and augmenting outcome rewards with process-level behavior regularization is proposed, which achieves the best performance among evaluated open-source RL-trained 8B models across four terminal-agent benchmarks.
Yi-Fan Yao, Bo Pang, Xuan-Phi Nguyen et al.· 2 citations
A Self-Evolutional single-agent/multi-agent Reinforcement Learning (SE-RL) framework that utilizes a Large Language Model (LLM) to design various RL algorithm modules, such as agent model design, reward function, profiling, communication, and state imagination, by leveraging the LLM generating module output or code.
Vincent Fu, Xin-Xin Xu, Weichen Xu et al.· Proceedings of the 32nd ACM...· 0 citations
UnifiedPlayers, a cooperative framework comprising a Planning Player that generates tasks, an Execution Player that produces multi-turn trajectories with Python tool calls, and an Evaluation Player that constructs executable verifiers, highlights cooperation among specialized players as a promising path toward self-enh...
This paper presents and evaluates a Hybrid AI Gaming Framework that runs three complementary modules asynchronously: a Proximal Policy Optimisation (PPO) reinforcement-learning agent for real-time difficulty calibration, a vector-grounded Memory Repository that anchors LLM-based NPC dialogue to the game's canonical wor...
Abhinav, Amandeep, Dharmender Kumar, Suraj, Keshav· International Journal of Adv...· 0 citations
These experiments provide empirical evidence that reasoning gains from gameplay can yield stronger transfer than math or code, and highlight GlyphBench's value as a testbed for systematic research on how language-model agents learn, interact, and generalize.
Roger Creus Castanyer, Marc-Alexandre Côté, Matthew J. Sargent et al.· 0 citations
Related blog posts
Microsoft Research Blog· microsoft.comNov 16, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 9, 2026
Assistant Professor Benjamin Lindquist studies how engineers deliberately injected randomness into early AI, giving rise to deep-learning models that sometimes hallucinate.