Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model
This work addresses the computational bottleneck of real-time planning for decision-making systems using world models, making them more practical for real-world deployment.
This paper introduces CompACT, a discrete tokenizer that compresses observations into just 8 tokens, significantly reducing the computational cost of planning in world models. This approach enables orders-of-magnitude faster planning while maintaining competitive performance.
World models provide a powerful framework for simulating environment dynamics conditioned on actions or instructions, enabling downstream tasks such as action planning or policy learning. Recent approaches leverage world models as learned simulators, but its application to decision-time planning remains computationally prohibitive for real-time control. A key bottleneck lies in latent representations: conventional tokenizers encode each observation into hundreds of tokens, making planning both slow and resource-intensive. To address this, we propose CompACT, a discrete tokenizer that compresses each observation into as few as 8 tokens, drastically reducing computational cost while preserving essential information for planning. An action-conditioned world model that occupies CompACT tokenizer achieves competitive planning performance with orders-of-magnitude faster planning, offering a practical step toward real-world deployment of world models.