Generative World Simulation with DAO and JING
DAO and JING couple a persistent shared world state with continuous first-person audiovisual generation.
The paper proposes a generative world-simulation framework pairing DAO, a computable world engine, with JING, an egocentric interactive experience model. DAO tracks characters, objects, spaces, and relationships and advances world time according to interaction rules. JING uses observer-specific conditions, references, history, and current actions to generate continuous first-person audiovisual experiences for humans and agents. Actions through viewpoint control and text instructions update both the shared state and subsequent experiences.
- DAO maintains persistent shared state for characters, objects, spaces, and relationships.
- JING generates each observer's first-person audiovisual experience from that state.
- User actions update the shared world and later generated experiences.
- The loop is meant to keep history beyond any single observer.
From the original article
1 min read at huggingface.co
We present a Generative World Simulation framework that couples persistent shared world state with continuous first-person audiovisual generation. The framework integrates DAO, a computable world engine, and JING, an Egocentric Interactive Experience Model, to support sustained interaction for humans and agents.
By XGEN Team, Guohao Cai, Haoxuan Che · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.