Newest papers

PrismWAM: Decoupling State Representations in World Action Models

Tianyu Xin, Wanqi Zhong, Bowen Hu, Zhangchi Guo, Lijie Wang, Yiming Wang, Haoran Liao, Heyang Yu, Xiaoyu Tian, Zhiding Yu, Yiming Li · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26564 · cited by 0 works (OpenAlex) · open-access PDF

Korean

세계 행동 모델 (WAMs)의 대규모 범위에서 심지어는 전공 된 비디오 모델의 잠재적 인 삽입을 통해 세계를 나타냅니다 (WAMs) 데이터는 일반적으로 넓은 범위에서 세계가 어떻게 보이는지 (예를 들어, 텍스트 및 색상)에서 물리적으로 구성되는 방법 (예를 들어, 점유율 및 포즈)로 나타납니다.이 목적은 $\textb{WAM}가 WAM의 상태를 구성해야한다는 미묘한 질문을 제기합니다. 우리는 세계 상태가 어떻게 나타나는지 (예를 들어, 텍스트 및 색상)에서 물리적으로 구성되는 방법을 분리해야한다고 주장합니다. 이전은 시멘틱 컨텍스트를 캡처하고, 후자는 물리적 상태에 의해 지배됩니다.이 목적을 위해,

Machine translation (Meta M2M100); the English original follows.

Abstract

World Action Models (WAMs) typically represent the world through latent embeddings from pretrained video models, which entangle heterogeneous aspects of the world, from how it looks (e.g., texture and color) to how it is physically configured (e.g., occupancy and pose). This raises an underexplored question: $\textbf{what should constitute the state of a WAM?}$ We argue that the world state should separate how the world appears from how the world evolves under action: the former captures semantic context, while the latter is governed by physical state. To this end, we propose $\textbf{PrismWAM}$, which decouples physics from semantics while jointly modeling state and action. PrismWAM factorizes an embodied task into a $\mathrm{Semantic\ Planning}\rightarrow\mathrm{Physical\ Rollout}\rightarrow\mathrm{Executable\ Action}$ pipeline: given only the initial RGB observation, semantic planning learns to predict the final goal image, while physical rollout predicts the fine-grained evolution toward this goal in depth space; executable actions are then generated from the depth rollout. Without any embodied action pretraining or additional data augmentation, PrismWAM achieves strong manipulation performance in both simulation and the real world. Notably, PrismWAM even surpasses a broad range of models with large-scale embodied action pretraining (e.g., OpenWAM-$α$). More broadly, we reframe World Action Models from modeling how the world looks to how it responds to action, paving the way for action-centric world models grounded in occupancy, flow, tactile sensing, proprioception, and beyond.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Action (physics), Embodied cognition, Semantics (computer science), Artificial intelligence, Decoupling (probability)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange