PhiZero：围绕“物理语言”构建的世界模型

AI 导读

PhiZero 是一种基于“物理语言”的物理世界模型，该语言通过自监督学习从野外视频中提取世界状态转移的紧凑离散表征。它采用先推理后渲染的范式，先以物理语言序列推断未来世界演化，再由扩散解码器渲染成视频。实验验证了其在物理一致性生成、细粒度动作条件模拟和零样本运动迁移上的能力。

推荐理由

这不是又一个视频生成模型，它把物理世界的变化压缩成一种「物理语言」再推演，让世界模型第一次有了显式的推理步骤。在物理一致性上甩开 SOTA 一截，做具身和机器人的值得细读。

正文

View PDF HTML (experimental)

Abstract:We introduce PhiZero, a physical world model built around physical language, a compact discrete representation of world-state transitions. Existing physical world models typically predict future videos directly in pixel space, leaving the underlying world dynamics implicit within high-dimensional visual predictors. Motivated by humans' ability to abstract predictive structure from visual experience and organize it in natural language for explicit reasoning, we learn physical language from in-the-wild videos through self-supervision and use it to explicitly reason about how the physical world evolves. Accordingly, PhiZero adopts a reason-then-render paradigm: it first infers future world evolution as a physical-language sequence and then renders the inferred transitions into videos. Extensive experiments across generation and understanding benchmarks validate the ability of PhiZero to model physically coherent world evolution. We further show its potential for realistic and interactive world modeling, fine-grained action-conditioned simulation, and zero-shot motion transfer.

Comments:	Project page: this https URL
Subjects:	Computer Vision and Pattern Recognition (cs.CV)
Cite as:	arXiv:2607.28624 [cs.CV]
	(or arXiv:2607.28624v1 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2607.28624 arXiv-issued DOI via DataCite

Submission history

From: Shuyao Shang [view email]
[v1] Thu, 30 Jul 2026 17:59:46 UTC (16,677 KB)