# MobileWAM: Bridging World Action Models to Mobile Manipulation with Chain-of-Foresight

arXiv: https://arxiv.org/abs/2608.04657 · HTML: https://mlsanity.com/paper/2608.04657/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Embodied and robot world models, one of 11 directions in World models.
- Builds on 18 of its 49 references that are in the map.
- First posted on arXiv 5 Aug 2026.

## Placed in

- World models / [Embodied and robot world models](https://mlsanity.com/wm/embodied-and-robot-world-models/)

## Builds on (most cited first, 18 of 18)

- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026)
- [Causal World Modeling for Robot Control](https://mlsanity.com/paper/2601.21998/) (2026)
- [ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI](https://mlsanity.com/paper/2410.00425/) (2024)
- [GigaWorld-Policy: An Efficient Action-Centered World--Action Model](https://mlsanity.com/paper/2603.17240/) (2026)
- [ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver](https://mlsanity.com/paper/2508.10333/) (2025)
- [ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?](https://mlsanity.com/paper/2606.19531/) (2026)
- [Next-Latent Prediction Transformers Learn Compact World Models](https://mlsanity.com/paper/2511.05963/) (2025)
- [ABot-M0.5: Unified Mobility-and-Manipulation World Action Model](https://mlsanity.com/paper/2607.00678/) (2026)
- [DualCoT-VLA: Visual-Linguistic Chain of Thought via Parallel Reasoning for Vision-Language-Action Models](https://mlsanity.com/paper/2603.22280/) (2026)
- [S-VAM: Shortcut Video-Action Model by Self-Distilling Geometric and Semantic Foresight](https://mlsanity.com/paper/2603.16195/) (2026)
- [EchoVLA: Robotic Vision-Language-Action Model with Synergistic Declarative Memory for Mobile Manipulation](https://mlsanity.com/paper/2511.18112/) (2025)
- [SG-VLA: Learning Spatially-Grounded Vision-Language-Action Models for Mobile Manipulation](https://mlsanity.com/paper/2603.22760/) (2026)
- [DSPv2: Improved Dense Policy for Effective and Generalizable Whole-body Mobile Manipulation](https://mlsanity.com/paper/2509.16063/) (2025)
- [MoMaGen: Generating Demonstrations under Soft and Hard Constraints for Multi-Step Bimanual Mobile Manipulation](https://mlsanity.com/paper/2510.18316/) (2025)
- [MobiGuide: Vocabulary-Guided Online Recovery for Mobile Manipulation](https://mlsanity.com/paper/2604.01567/) (2026)
- [World-Ego Modeling for Long-Horizon Evolution in Hybrid Embodied Tasks](https://mlsanity.com/paper/2605.19957/) (2026)

## Built on by (most cited first, 3 of 3)

- [MoPA: Coordinated Mobile Manipulation via Subsystem-Specific Perception Alignment](https://mlsanity.com/paper/2609.12081/) (2026)
- [UniWAM: Unified World-Action Model](https://mlsanity.com/paper/2610.02054/) (2026)
- [UniWAM Technical Report: Unified Mobile Manipulation via Mixed-Stream World-Action Modeling and Manipulation Anchor Pose Supervision](https://mlsanity.com/paper/2609.39388/) (2026)

## Abstract (arXiv)

World action models (WAMs) built on video generation backbones are a rising recipe for robot learning, yet remain confined to tabletop manipulation. Mobile manipulation demands simultaneous locomotion and whole-body manipulation amid scene-scale dynamics, yet is still dominated by dynamics-blind visual encoders with hand-crafted coordination. We bridge this gap with MobileWAM, a mixture-of-transformers architecture that fuses a pretrained video diffusion transformer with a lightweight action expert through layerwise joint attention, translating internet-scale motion priors into whole-body control. To reconcile the heterogeneous dynamics of moving and manipulating, each feed-forward layer of the action expert becomes a three-expert mixture of shared, locomotion, and manipulation experts, softly routed by the motion intent in the action tokens. To densify supervision, we further propose Chain-of-Foresight (CoF): intermediate representations sequentially predict a chain of future latent chunks, each step conditioned on its predecessor. CoF pairs naturally with our decoupled video--action denoising scheme. At deployment, the WAM serves as a pure current-frame encoder; foresight acts only through gradients, so at inference the foresight chain and video generation are discarded, leaving only policy-level cost. MobileWAM surpasses state-of-the-art mobile manipulation policies on ManiSkill-HAB and fine-tunes to a real ARX Lift2 mobile manipulator across diverse tasks with strong generalization. Code will be released soon.