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 · 2026 · arXiv

DOI: 10.48550/arXiv.2610.26564 · open-access PDF

Indonesian

World Action Models (WAM) biasanya mewakili dunia melalui embeddings latent dari model video pretrained, yang membingungkan aspek heterogen dari dunia, dari bagaimana ia terlihat (misalnya, tekstur dan warna) ke bagaimana ia dikonfigurasi secara fisik (misalnya, penempatan dan pose). Ini menimbulkan pertanyaan yang belum dijelaskan: $\textbf{apa yang harus membentuk keadaan WAM?}$ Kami berpendapat bahwa negara dunia harus memisahkan bagaimana dunia terlihat dari bagaimana dunia berkembang di bawah tindakan: yang pertama menangkap konteks semantik, sementara yang terakhir dikendalikan oleh keadaan fisik.

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

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

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