# RoboStereo: Dual-Tower 4D Embodied World Models for Unified Policy Optimization

arXiv: https://arxiv.org/abs/2603.12639 · HTML: https://mlsanity.com/paper/2603.12639/

## Key facts

- Cited by 10 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 28 of its 52 references that are in the map.
- First posted on arXiv 13 Mar 2026.

## Placed in

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

## Builds on (most cited first, 28 of 28)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection](https://mlsanity.com/paper/2203.03605/) (2022)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [Depth Anything 3: Recovering the Visual Space from Any Views](https://mlsanity.com/paper/2511.10647/) (2025)
- [Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation](https://mlsanity.com/paper/2508.05635/) (2025)
- [TesserAct: Learning 4D Embodied World Models](https://mlsanity.com/paper/2504.20995/) (2025)
- [Ovi: Twin Backbone Cross-Modal Fusion for Audio-Video Generation](https://mlsanity.com/paper/2510.01284/) (2025)
- [WMPO: World Model-based Policy Optimization for Vision-Language-Action Models](https://mlsanity.com/paper/2511.09515/) (2025)
- [WoW: Towards a World omniscient World model Through Embodied Interaction](https://mlsanity.com/paper/2509.22642/) (2025)
- [VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulators](https://mlsanity.com/paper/2510.00406/) (2025)
- [GigaWorld-0: World Models as Data Engine to Empower Embodied AI](https://mlsanity.com/paper/2511.19861/) (2025)
- [Igniting VLMs toward the Embodied Space](https://mlsanity.com/paper/2509.11766/) (2025)
- [TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimization](https://mlsanity.com/paper/2506.08440/) (2025)
- [Learning Video Generation for Robotic Manipulation with Collaborative Trajectory Control](https://mlsanity.com/paper/2506.01943/) (2025)
- [SAM-R1: Leveraging SAM for Reward Feedback in Multimodal Segmentation via Reinforcement Learning](https://mlsanity.com/paper/2505.22596/) (2025)
- [WristWorld: Generating Wrist-Views via 4D World Models for Robotic Manipulation](https://mlsanity.com/paper/2510.07313/) (2025)
- [Vidarc: Embodied Video Diffusion Model for Closed-loop Control](https://mlsanity.com/paper/2512.17661/) (2025)
- [Alignment is All You Need: A Training-free Augmentation Strategy for Pose-guided Video Generation](https://mlsanity.com/paper/2408.16506/) (2024)
- [Zo3T: Zero-Shot 3D-Aware Trajectory-Guided Image-to-Video Generation via Test-Time Training](https://mlsanity.com/paper/2509.06723/) (2025)
- [MIND-V: Hierarchical World Model for Long-Horizon Robotic Manipulation with RL-based Physical Alignment](https://mlsanity.com/paper/2512.06628/) (2025)
- [One4D: Unified 4D Generation and Reconstruction via Decoupled LoRA Control](https://mlsanity.com/paper/2511.18922/) (2025)
- [Inferix: A Block-Diffusion based Next-Generation Inference Engine for World Simulation](https://mlsanity.com/paper/2511.20714/) (2025)

## Built on by (most cited first, 10 of 10)

- [ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?](https://mlsanity.com/paper/2606.19531/) (2026)
- [SolarWM: Open Data and Scalable Training for Long-Horizon Video World Models](https://mlsanity.com/paper/2609.02886/) (2026)
- [KVPO: ODE-Native GRPO for Autoregressive Video Alignment via KV Semantic Exploration](https://mlsanity.com/paper/2605.14278/) (2026)
- [A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model](https://mlsanity.com/paper/2604.05672/) (2026)
- [When Should the Teacher Move? Temporal Coupling and Stability in Self On-Policy Distillation](https://mlsanity.com/paper/2606.03532/) (2026)
- [PhysEditWorld: A Large-Scale Dataset Toward Physics-Editable World Models](https://mlsanity.com/paper/2606.26694/) (2026)
- [WorldCast: Distributed Multiplayer World Models](https://mlsanity.com/paper/2610.12412/) (2026)
- [SimForcing: Distilling Simulation Motion Priors into Real-Domain Robot World Models](https://mlsanity.com/paper/2610.06598/) (2026)
- [Memory Forcing: Attendable Mid-Horizon History for Streaming Video Generation](https://mlsanity.com/paper/2610.11756/) (2026)
- [ContourVLA: A Closed-Loop Perception-Action Contour Policy for Generalized Referring Expression Segmentation](https://mlsanity.com/paper/2610.12107/) (2026)

## Abstract (arXiv)

Scalable Embodied AI faces fundamental constraints due to prohibitive costs and safety risks of real-world interaction. While Embodied World Models (EWMs) offer promise through imagined rollouts, existing approaches suffer from geometric hallucinations and lack unified optimization frameworks for practical policy improvement. We introduce RoboStereo, a symmetric dual-tower 4D world model that employs bidirectional cross-modal enhancement to ensure spatiotemporal geometric consistency and alleviate physics hallucinations. Building upon this high-fidelity 4D simulator, we present the first unified framework for world-model-based policy optimization: (1) Test-Time Policy Augmentation (TTPA) for pre-execution verification, (2) Imitative-Evolutionary Policy Learning (IEPL) leveraging visual perceptual rewards to learn from expert demonstrations, and (3) Open-Exploration Policy Learning (OEPL) enabling autonomous skill discovery and self-correction. Comprehensive experiments demonstrate RoboStereo achieves state-of-the-art generation quality, with our unified framework delivering >97% average relative improvement on fine-grained manipulation tasks.