Skip to content

DiffWAM: A Fast and Efficient Navigation World Action Model

Sep 2026 · 1 citation · 36 references
Computer Science

TL;DR

DiffWAM is presented, a geometry-conditioned navigation world-action model that directly transforms multi-level predictive features into continuous camera trajectories, providing a direct alternative to generate-then-reconstruct navigation pipelines.

Abstract

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/.

View source

Similar papers

Preprint Aug 2026

GaussianWAM: Distilling Geometry and Semantics from 3D Gaussian Fields into World-Action Models

GaussianWAM is proposed, a training-time representation-enhancement framework that organizes geometric and semantic supervision through a 3D Gaussian field and improves performance on standard LIBERO and shows positive transfer trends on RoboTwin and real-world manipulation.

Zi-Jian Zhang, Yu-Qing Jiang, Wei-Tao Zhou et al. · 3 citations
Preprint Aug 2026

Revisiting Local Context for Long-Horizon Streaming 3D Reconstruction

ABot-Recon is presented, a simple streaming model that caches KV features from only the preceding 11 frames and predicts a point map in the current camera coordinate system together with an adjacent-frame relative pose that remains equivariant under changes of reference frame.

Jiarong Han, Jincheng Xiong, Yu-Zhou Liu et al. · 2 citations · ⚡1
Sep 2026

MVG-WAM: Multiple View Geometry-Aware World-Action Modeling for Robotic Manipulation

World-Action Models (WAMs) couple visual dynamics with action prediction, bringing the rich priors of pretrained video models to robotic manipulation. However, their multi-view interfaces typically tile images or concatenate tokens, leaving the geometric relationships among synchronized cameras implicit. This makes it...

Wen-Bo Chen, Tian-Fu Li, Hao-Xuan Xu et al. · 1 citation
Preprint Sep 2026

Seeing the World and the Self from Egocentric Video

Experiments show that RESELF outperforms state-of-the-art methods designed for the individual tasks across depth estimation, camera tracking, and full-body motion estimation.

Kai Guan, Minchao Jiang, Ruichen WangLi et al. · 0 citations
Preprint Sep 2026

GeoLAM: Learning Geometry-Grounded Latent Actions from Unlabeled Human Videos

GeoLAM, a framework for learning geometry-grounded latent actions from action-free human videos, combines future-frame reconstruction through a frozen geometric feature hierarchy with motion supervision from a training-only 4D geometry teacher and requires neither the geometry teacher nor future-video generation.

Yi-Fan Xie, He-Kun Tian, Jin-Kun Liu et al. · 0 citations
Preprint Aug 2026

GaussianDream++: Efficient 3D Gaussian World Modeling for Robotic Manipulation

GaussianDream demonstrates that training-time current Gaussian reconstruction and future Gaussian prediction provide effective 3D supervision, but its dense VGGT/TGE-based prefix jointly carries state, dynamics, and action-conditioning information.

Yu-Qing Jiang, Zi-Jian Zhang, Wei-Tao Zhou et al. · 1 citation

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.