Summary
A new arXiv survey (2610.31668) by Shengqiong Wu, Hao Fei, and colleagues including Ziwei Liu, Shuicheng Yan, Ming-Hsuan Yang, and Philip Torr examines video foundation models and video world models through the unifying lens of world modeling. The authors treat observed video as a partial, lossy, temporally ordered projection of a latent temporal world and organize the literature around four complementary operations: reading (inferring world properties from observed video), writing (synthesizing video from specified or learned world conditions), sharing (integrating understanding and generation through common modeling mechanisms), and interacting (predicting and updating world states under actions and evolving observations). This perspective distinguishes temporal consistency within sequences from consistency of the underlying world state. The survey traces a shift from clip-level recognition and perceptual fidelity toward persistent state, controllable dynamics, physical and causal validity, and interaction- and decision-driven evaluation, and distills resulting open challenges in computer vision.
Paper Overview
Field: Computer Vision (CV)
Authors: Shengqiong Wu, Meng Luo, Bobo Li, Ziwei Liu, Shuicheng Yan, Ming-Hsuan Yang, Philip Torr, Hao Fei
Published: 2026-10-22
arXiv: 2610.31668
Abstract
Video foundation models have advanced rapidly in understanding and generating visual sequences, while video world models seek to capture the underlying dynamics of the world they depict. Despite their different objectives, both concern the relationship between video observations and an evolving world. This survey adopts world modeling as a unifying lens for examining these developments.
Viewing an observed video as a partial, lossy, and temporally ordered projection of a latent temporal world, the authors organize the literature around four complementary operations:
- Reading — infers world properties from observed video;
- Writing — synthesizes video from specified or learned world conditions;
- Sharing — integrates understanding and generation through common modeling mechanisms;
- Interacting — predicts and updates world states under actions and evolving observations.
This perspective distinguishes temporal consistency within observed or generated sequences from consistency of the underlying world state. The survey traces the field's evolution from clip-level recognition and perceptual fidelity toward persistent state modeling, controllable dynamics, physical and causal validity, and evaluation based on interaction and downstream decision-making, and distills the open challenges that follow.
---
*Auto-collected on 2026-10-24*
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/178636401