Wolfgang Lehrach | Code World Models for General Game Playing
While Large Language Models (LLMs) show promise in many domains, relying on them for direct policy generation in games often results in illegal moves and poor strategic play.
Get the latest news, advances in research, policy work, and education program updates from HAI in your inbox weekly.
Sign Up For Latest News
While Large Language Models (LLMs) show promise in many domains, relying on them for direct policy generation in games often results in illegal moves and poor strategic play.
Strategic stability exists when neither side thinks it can improve its strategic outcome by striking first.

Strategic stability exists when neither side thinks it can improve its strategic outcome by striking first.
The 2026 RAISE Health Symposium will convene leading voices in technology, medicine, and policy to explore what’s real, what’s next, and what’s needed to ensure AI advances responsibly and benefits all. Registration opens March 11.

The 2026 RAISE Health Symposium will convene leading voices in technology, medicine, and policy to explore what’s real, what’s next, and what’s needed to ensure AI advances responsibly and benefits all. Registration opens March 11.
Music is intertwined with human emotion, memory, and identity, making it a powerful medium for affective experience and regulation.

Music is intertwined with human emotion, memory, and identity, making it a powerful medium for affective experience and regulation.
In this talk, I present an approach that moves away from direct prompting, instead using LLMs as program synthesizers to bridge the gap between natural language rules and symbolic world models. The LLM receives a game description and example trajectories, and outputs an executable, symbolic world model (CWM) represented in Python. The trajectories also ensure the rules are correctly captured and aid in refining the CWM if they are not. Note that even trajectories containing only a single player's observations and actions can be used to help validate and refine CWMs. Furthermore, partially observed trajectories also allow comparisons between CWMs via a bound on the likelihood.
Given a CWM, Monte Carlo Tree Search (MCTS) or Reinforcement Learning (RL) methods can play the game, and gameplay can be further enhanced by adding in LLM-derived synthesized value functions. Imperfect information games are handled by having the LLM synthesize inference functions to impute information sets, or by directly training reinforcement learning policies on top of the CWM.