OpenNWM: A Generalist Navigation World Model with Latent Action Pretraining
cs.RO, cs.AI
Submitted: 2026-11-01
Updated: 2026-11-05
Project page: https://openspace-lab.github.io/OpenNWM/ABSTRACT
Terminology
Sources
- EgoWalk: A Multimodal Dataset for Robot Navigation in the Wild
- Navigation World Models
- Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets
- villa-X: Enhancing Latent Action Modeling in Vision-Language-Action Models
- AdaWorld: Learning Adaptable World Models with Latent Actions
- DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos
- Vid2World: Crafting Video Diffusion Models to Interactive World Models
- Olaf-World: Orienting Latent Actions for Video World Modeling
- Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model
- LatBot: Distilling Universal Latent Actions for Vision-Language-Action Models
- From Pixels to Tokens: A Systematic Study of Latent Action Supervision for Vision-Language-Action Models
- DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision
- LARA: Latent Action Representation Alignment for Vision-Language-Action Models
- HumanScale: Egocentric Human Video Can Outperform Real-Robot Data for Embodied Pretraining
- Latent Action Learning Requires Supervision in the Presence of Distractors
- Learning to Act without Actions
- Rapid Exploration for Open-World Navigation with Latent Goal Models
- DINOv3
- SpikeVLA: Vision-Language-Action Models with Spiking Neural Networks
- i2Nav-Robot: A Large-Scale Indoor-Outdoor Robot Dataset for Multi-Sensor Fusion Navigation
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation