RIWANav: Recursive World-Action Models with Self-Improvement for Urban Navigation
cs.RO
Submitted: 2026-10-06
Updated: 2026-10-06
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- World Action Models are Zero-shot Policies
- Fast-WAM: Do World Action Models Need Test-time Future Imagination?
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration
- OmniVLA: An Omni-Modal Vision-Language-Action Model for Robot Navigation
- VENTURA: Adapting Image Diffusion Models for Unified Task Conditioned Navigation
- Cosmos 3: Omnimodal World Models for Physical AI
- Proximal Policy Optimization Algorithms
- On Evaluation of Embodied Navigation Agents
- General Evaluation for Instruction Conditioned Navigation using Dynamic Time Warping
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