XGenAct: Geometry-Enhanced World Action Models through Cross-Task Generation
cs.RO, cs.CV, cs.LG
Submitted: 2026-10-02
Updated: 2026-10-02
Comments: 27 pages, including appendix
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- A Power Transform
- LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policies
- FlowWAM: Optical Flow as a Unified Action Representation for World Action Models
- Dream2Flow: Bridging Video Generation and Open-World Manipulation with 3D Object Flow
- Image Generators are Generalist Vision Learners
- RT-Trajectory: Robotic Task Generalization via Hindsight Trajectory Sketches
- Unified 4D World Action Modeling from Video Priors with Asynchronous Denoising
- MolmoAct: Action Reasoning Models that can Reason in Space
- WAM4D: Fast 4D World Action Model via Spatial Register Tokens
- GeoSem-WAM: Geometry- and Semantic-Aware World Action Models
- PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- THE COLOSSEUM: A Benchmark for Evaluating Generalization for Robotic Manipulation
- Wan: Open and Advanced Large-Scale Video Generative Models
- Video Generation Models are General-Purpose Vision Learners
- World Action Models are Zero-shot Policies
- DreamWAM: Beyond RGB Future Prediction for World Action Models
- SUV: Future Scene Understanding as Video Generation for End-to-End Driving
- DICEPTION: A Generalist Diffusion Model for Visual Perceptual Tasks
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies