Reading, Writing, Sharing, and Interacting with the World through Video

One-line takeaway

This survey unifies video understanding and generation through world modeling, focusing on four operations: reading, writing, sharing, and interaction, shifting from clip-level recognition to persistent, causal state modeling.

论文从世界建模视角系统梳理了视频理解与生成的四大核心操作：阅读、写作、共享和交互，强调从片段识别转向对持久状态和因果一致性的建模。

Key points

This survey presents a unifying framework for video foundation models through world modeling, organizing the literature around four operations: reading, writing, sharing, and interacting. It highlights a shift from clip-level perception to persistent world states, controllable evolution, and causal validity, emphasizing evaluation via interaction and downstream decisions.

• Proposes world modeling as a unifying lens for video understanding and generation.
• Organizes literature into four operations: reading, writing, sharing, and interaction.
• Shifts emphasis from clip-level recognition to persistent, causally valid world states.
• Highlights evaluation via interaction and downstream decision-making as key for validity.

Written only from the official arXiv abstract by an open-source model, then checked automatically (every number and name must appear in the abstract). The paper is authoritative.