# DiffWAM: A Fast and Efficient Navigation World Action Model

arXiv: https://arxiv.org/abs/2609.39763 · HTML: https://mlsanity.com/paper/2609.39763/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in World models for navigation, one of 11 directions in World models.
- Builds on 17 of its 36 references that are in the map.
- First posted on arXiv 30 Sep 2026.

## Placed in

- World models / [World models for navigation](https://mlsanity.com/wm/world-models-for-navigation/)

## Builds on (most cited first, 17 of 17)

- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [World Action Models are Zero-shot Policies](https://mlsanity.com/paper/2602.15922/) (2026)
- [Policy Distillation](https://mlsanity.com/paper/1511.06295/) (2015)
- [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026)
- [π3: Permutation-Equivariant Visual Geometry Learning](https://mlsanity.com/paper/2507.13347/) (2025)
- [VPO: Aligning Text-to-Video Generation Models with Prompt Optimization](https://mlsanity.com/paper/2503.20491/) (2025)
- [VLA-AN: An Efficient and Onboard Vision-Language-Action Framework for Aerial Navigation in Complex Environments](https://mlsanity.com/paper/2512.15258/) (2025)
- [Faster-WAM: Efficient Inference-Time Future Conditioning for Robust World Action Models](https://mlsanity.com/paper/2608.04404/) (2026)
- [WorldVLN: Autoregressive World Action Model for Aerial Vision-Language Navigation](https://mlsanity.com/paper/2605.15964/) (2026)
- [ImagineUAV: Aerial Vision-Language Navigation via World-Action Modeling and Kinodynamic Planning](https://mlsanity.com/paper/2606.01205/) (2026)
- [WorldFly: A World-Model-Based Vision-Language-Action Model for UAV Navigation](https://mlsanity.com/paper/2606.06147/) (2026)
- [Scaling Laws for Speculative Decoding](https://mlsanity.com/paper/2505.07858/) (2025)
- [FlyMirage: A Fully Automated Generation Pipeline for Diverse and Scalable UAV Flight Data via Generative World Model](https://mlsanity.com/paper/2605.19600/) (2026)
- [PathPainter: Transferring the Generalization Ability of Image Generation Models to Embodied Navigation](https://mlsanity.com/paper/2605.07496/) (2026)
- [Towards Precise Intent-Aligned VLA Aerial Navigation via Expert-Guided GRPO](https://mlsanity.com/paper/2606.02313/) (2026)
- [NavGen: Visual Generative Models as a Scalable Data Engine for Embodied 3D Navigation](https://mlsanity.com/paper/2609.30770/) (2026)

## Abstract (arXiv)

Pretrained video foundation models encode rich semantic and spatiotemporal priors for embodied navigation, yet converting these priors into UAV motion typically requires expensive future-video synthesis and geometric reconstruction. We investigate whether the motion implicit in future visual prediction can instead be recovered directly from the predictive representations of a frozen video model. To this end, we present DiffWAM, a geometry-conditioned navigation world-action model that directly transforms multi-level predictive features into continuous camera trajectories. Its Grid-Motion module preserves spatial-temporal motion associations, while Latent2Pose grounds them with first-frame geometry to recover metrically meaningful 3D motion. Complete video rollouts and geometric reconstruction are required only for offline supervision, eliminating future-video decoding and multi-frame reconstruction during deployment. We further introduce FastDreamer, which overlaps predictive and geometric computation with ongoing flight and performs timestamp-aware asynchronous trajectory handoff for continuous UAV execution. DiffWAM achieves a trajectory RMSE of 0.3492 m and an endpoint success rate of 74.40% on the 1,000-sample DiffWAM-1000 benchmark, while representative real-world experiments demonstrate complex behaviors including constrained traversal, orbiting, S-shaped flight, and multi-stage navigation. An onboard DiffWAM-Flash implementation further reaches 1.08 s model-pipeline latency on NVIDIA Jetson AGX Thor. These results demonstrate that predictive video representations can be efficiently grounded into continuous 3D motion, providing a direct alternative to generate-then-reconstruct navigation pipelines. Project page: https://zzmmzzm.github.io/diffwam.github.io/.