Article

Controlling Test-Time Compute for World-Model Planning with Expected Free Energy: When to Stop Thinking

The study examines whether expected free energy (EFE) can determine how much computation to allocate to planning based on a learned world model, with an objective combining goal distance and disagreement among stochastic predictions. Experiments found that stopping when elite cost stalls reduces refinement while mean success rates stay close to fixed-depth references; with a frozen model, rewarding disagreement lowers success, while online updates partially recover losses caused by regions missing from the training data. The cost-stall rule fails when optimization exploits model errors, and predicted cost keeps falling after task success rates peak.

Part of the event: An Experimental Study on EFE-Controlled Planning Compute in World Models

This page provides a summary and source link only.