通过循环架构提升世界模型的长时序模拟效率，节省100倍参数。
Looped World Models
- 用共享参数的Transformer循环迭代优化潜在环境状态。
- 相比传统方法参数效率提升最高达100倍，且计算深度可自适应调整。
- 适合追求高效长时序建模的研究者与部署场景受限的应用开发。
当前世界模型面临根本矛盾：高保真长时序模拟需要深度计算，但深层模型部署成本高且误差易累积。我们提出首个循环式世界模型架构——Looped World Models（LoopWM），通过参数共享的Transformer块迭代精炼潜在环境状态。该方法在保持性能的同时，相较传统方案实现高达100倍的参数效率，并支持按预测步骤复杂度自动调节计算深度。不同于单纯扩大模型规模或训练数据，LoopWM将迭代潜空间深度确立为世界模拟的新扩展维度，有望显著推动该领域发展。
原文摘要 · Abstract (English)
Current world models face a fundamental tension: faithful long-horizon simulation demands deep computation, but deeper models are expensive to deploy and prone to compounding errors. We resolve this by introducing Looped World Models (LoopWM), which are the first looped architectures for world modelling. Our method iteratively refines latent environment states through a parameter-shared transformer block. This yield up to 100x parameter efficiency over conventional approaches with adaptive computation that automatically scales depth to match the complexity of each prediction step. Orthogonal to scaling model size and training data, LoopWM establishes iterative latent depth as a new scaling axis for world simulation, which might significantly push the community forward.
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。