用8个离散标记压缩观测，让世界模型实时规划成为可能
Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model
- 设计紧凑离散分词器CompACT，每帧图像仅需8个标记
- 在相同精度下，规划速度提升数量级，内存占用大幅降低
- 适合需要快速决策的机器人控制、自动驾驶等场景
世界模型通过条件化动作或指令模拟环境动态，支持动作规划与策略学习等下游任务。现有方法虽将世界模型作为可学习的模拟器，但在决策时规划仍因计算成本过高难以实现实时控制。关键瓶颈在于潜在表示：传统分词器将每个观测编码为数百个标记，导致规划效率低下且资源消耗大。为此，我们提出CompACT，一种离散分词器，可将每个观测压缩至仅8个标记，显著降低计算开销的同时保留规划所需的关键信息。采用CompACT的行动条件世界模型在规划性能上达到相当水平，但规划速度提升数量级，为世界模型在真实场景中的部署提供了实用路径。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。