video world model
A video world model is a system that generates video by keeping track of what’s happening in the scene over time, like a mental map that remembers where objects are, how they move, and even things that are temporarily out of sight. Unlike simple video generators that only create frames one by one, a world model maintains consistency, so if the camera looks away and then returns, the same objects are still there, possibly having changed as expected.
Across the papers, the common thread is improving how video world models represent and control the persistent state of a scene. They do this by adding explicit 3D or 4D representations (like object meshes, depth maps, or abstract state tables), by separating visual generation from world-state tracking, and by creating specialized datasets and benchmarks to test long-term consistency and controllability. The focus is on enabling complex interactions, such as moving the camera, changing weather, or editing objects, while keeping the generated world coherent across many frames.
- Papers
- 6
- Released code
- 0
- First seen
- Oct 2026
- Latest
- Oct 2026
6 papers in the last two months, against 0 in the two before.
Start here
Three papers to read first, in the order they came out- 1Symmetry-regularized flow matching enables video generation from any camera pose without ground truth novel views
SymRegFlow generates multi-view consistent videos across continuous viewpoints without novel-view RGB supervision, using masked dual-anchor supervision and cross-anchor denoising consistency.
Oct 2026 - 2Decoupling abstract state from visual state improves video world model consistency
StateCraft pairs a language model with a video diffusion model to maintain a persistent abstract state for consistent and controllable video world simulation.
Oct 2026 - 3Plane primitives serve as compact persistent memory for video world models
Plane primitives can serve as a compact persistent 3D memory for video world models, enabling coherent long-context generation.
Oct 2026
How fast it is moving
Often studied with
diffusion modelA type of generative model that learns to reverse a gradual noising process to create data.3 togetherflow matchingA generative training method that learns an ordinary differential equation mapping noise to data by matching velocity fields along probability paths.2 togetherdiffusion transformerA neural network architecture combining transformer layers with diffusion processes for high-quality image and video generation.1 togetherWho is working on it
The papers
Most central to this idea first, not most recent.
- Chinese Techcs.CV
Plane primitives serve as compact persistent memory for video world models
Plane primitives can serve as a compact persistent 3D memory for video world models, enabling coherent long-context generation.
Tsinghua University, Ant Group · Oct 2026
- Big Techcs.CV
Decoupling abstract state from visual state improves video world model consistency
StateCraft pairs a language model with a video diffusion model to maintain a persistent abstract state for consistent and controllable video world simulation.
INSAIT / Sofia University, ETH Zurich · Oct 2026
- Top Universitycs.CV
Symmetry-regularized flow matching enables video generation from any camera pose without ground truth novel views
SymRegFlow generates multi-view consistent videos across continuous viewpoints without novel-view RGB supervision, using masked dual-anchor supervision and cross-anchor denoising consistency.
Tsinghua University, Robert Bosch GmbH · Oct 2026
- Chinese Techcs.CV
Explicit weather transitions improve control in generated future-scene videos
MeteoVerse generates future videos from a single image with explicit control over weather transitions including introduction, preservation, and removal.
Harbin Institute of Technology, Huawei · Oct 2026
- AI Startupcs.CV
Complete 3D meshes improve control of generated video motion
Using explicit rigid 3D geometry to control video world models achieves better camera and object control than prior methods.
Stability AI, University of Illinois Urbana-Champaign · Oct 2026
- Chinese Techcs.CV
Benchmark exposes separate static and dynamic memory failures in video worlds
WorldMemBench evaluates video world models on static and dynamic memory retention across repeated revisits, finding no model leads all dimensions.
Zhejiang University, Ant Group · Oct 2026
Concepts are extracted from each paper and reused across the corpus, so this page grows on its own as the desk reads.