WorldAlign introduces a decoupled 4D reward framework for improving world consistency in generated videos. It separates static regions from dynamic subjects, aligning each with a suitable world prior—geometric priors for static geometry and a vision‑language model for dynamic subjects—using masked reprojection and a VLM‑as‑a‑judge reward. The method enhances both static and dynamic consistency in pretrained image‑to‑video generators without relying on human preference data.
arXiv:2610.14293v1 Announce Type: cross Abstract: Video diffusion models increasingly serve as priors for 4D generation, but the generated videos often lack geometric consistency. Recent approaches a...
arXiv:2605. 16713v2 Announce Type: replace-cross Abstract: Modern Vision-Language Models (VLMs) achieve strong semantic recognition, yet remain brittle on elementary spatial relations such as left of, on, behind, and between.
WorldReward introduces a vision‑language model–based reward system for camera‑conditioned world models, combining action consistency and visual quality evaluation. It processes paired videos by splitting them into action‑aligned chunks, structuring visual evidence, and aggregating decisions through voting. The model is trained on a large, reasoning‑augmented preference dataset and outperforms GPT‑5.5 on a human‑annotated benchmark, improving both action execution and visual quality when applied to RL post‑training.
arXiv:2610.02160v1 Announce Type: new Abstract: Precise control over camera and object motion is essential for professional video production. Existing methods control objects only coarsely, through i...
Orbia is a new benchmark that systematically evaluates 3D consistency in interactive video world models. It organizes evaluation into three dimensions—input preservation, generated‑content persistence, and 3D self‑consistency—using controlled exploration and revisit protocols. The benchmark covers 720 instances across indoor and outdoor scenes and tests 27 different models, providing a comprehensive framework for diagnosing and improving 3D consistency.
arXiv:2606. 02753v1 Announce Type: cross Abstract: Video world models are a foundational generative technology for embodied AI and the Metaverse, yet existing approaches are inherently limited to a single agent observing from a single perspective.
World in World introduces a training‑free, inference‑time interface that transforms diverse control signals—such as source‑video observations, target‑view projections, geometry renderings, and retrieved states—into camera‑ and time‑labelled visual states. These states are processed by a frozen causal video model’s self‑attention, enabling tasks like camera‑controlled rerendering, long‑horizon revisiting, and human‑motion transfer without additional training. The method employs a correspondence router and evidence‑wise attention to align token identities and regulate auxiliary channel contributions during a single denoising pass.
arXiv:2610.07031v1 Announce Type: new Abstract: Recent video world models have witnessed the paradigm shift from single-agent to multi-agent involvements, which can reveal more complicated dynamics a...
4D dynamic scene understanding requires grounding language to a persistent worldline that binds identity, metric 3D motion, and synchronized multi-view 2D projections. Existing paradigms capture only part of this structure: large multimodal models reason over rich visual evidence but rarely preserve metric topology, while vision-language tracking remains tied to fragmented 2D or 3D outputs and local continuation.
LoGo introduces a hybrid reward system combining global and spatially localized rewards to improve 3D consistency in camera-controlled video generation. The local reward offers fine-grained credit assignment, reducing object shifts and artifacts, while the global reward maintains camera following and overall video quality. Evaluations on DL3DV and the new TrajectoryBench benchmark demonstrate LoGo’s advantage over existing post-training techniques.
Reconstructing humans and their surrounding environments in a globally consistent 4D space is essential for comprehensive perception. However, prior works typically assume single-view inputs or decouple humans, scenes, and cameras, making them unable to recover coherent geometry, stable motion, and physically aligned trajectories.
SolarWM is an open foundation for building interactive video world models, offering a reconfigurable multi‑source data engine that unifies 1.43 million clips from 10 datasets into a consistent, frame‑aligned format. It provides a backbone‑native adaptation framework that preserves native representations of models ranging from 5 B to 33 B parameters, and a three‑stage training recipe combining bidirectional adaptation, teacher‑forced autoregressive initialization, and distribution‑matching distillation. The resulting causal models can interact in real‑time over rollouts from minutes to hours, trained only on 5‑second sequences, and the project releases data, pipeline, recipes, weights, and framework for reproducible research.