# RDT2: Exploring the Scaling Limit of UMI Data Towards Zero-Shot Cross-Embodiment Generalization

arXiv: https://arxiv.org/abs/2602.03310 · HTML: https://mlsanity.com/paper/2602.03310/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- Placed in Generalist VLA models and benchmarks, one of 13 directions in Vision-language-action models.
- Builds on 68 of its 76 references that are in the map.
- First posted on arXiv 3 Feb 2026.

## Placed in

- Vision-language-action models / [Generalist VLA models and benchmarks](https://mlsanity.com/vla/generalist-vla-models-and-benchmarks/)

## Builds on (most cited first, 30 of 68)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [Generating Diverse High-Fidelity Images with VQ-VAE-2](https://mlsanity.com/paper/1906.00446/) (2019)
- [GR00T N1: An Open Foundation Model for Generalist Humanoid Robots](https://mlsanity.com/paper/2503.14734/) (2025)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [Open X-Embodiment: Robotic Learning Datasets and RT-X Models](https://mlsanity.com/paper/2310.08864/) (2023)
- [DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset](https://mlsanity.com/paper/2403.12945/) (2024)
- [Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs](https://mlsanity.com/paper/2406.16860/) (2024)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Autoregressive Image Generation using Residual Quantization](https://mlsanity.com/paper/2203.01941/) (2022)
- [RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation](https://mlsanity.com/paper/2410.07864/) (2024)
- [Vector-quantized Image Modeling with Improved VQGAN](https://mlsanity.com/paper/2110.04627/) (2021)
- [BridgeData V2: A Dataset for Robot Learning at Scale](https://mlsanity.com/paper/2308.12952/) (2023)

## Built on by (most cited first, 30 of 30)

- [ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning](https://mlsanity.com/paper/2602.11236/) (2026)
- [World Action Models: The Next Frontier in Embodied AI](https://mlsanity.com/paper/2605.12090/) (2026)
- [Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Execution](https://mlsanity.com/paper/2602.12684/) (2026)
- [Xiaomi-Robotics-1: Scaling Vision-Language-Action Models with over 100K Hours of Real-World Trajectories](https://mlsanity.com/paper/2607.15330/) (2026)
- [EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policies](https://mlsanity.com/paper/2606.20092/) (2026)
- [FASTER: Rethinking Real-Time Flow VLAs](https://mlsanity.com/paper/2603.19199/) (2026)
- [FRAPPE: Infusing World Modeling into Generalist Policies via Multiple Future Representation Alignment](https://mlsanity.com/paper/2602.17259/) (2026)
- [LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfer](https://mlsanity.com/paper/2602.10556/) (2026)
- [ACE-Ego-0: Unifying Egocentric Human and Robotic Data for VLA Pretraining](https://mlsanity.com/paper/2606.17200/) (2026)
- [Motus2: A Self-Evolving General World Model for Dexterous Manipulation](https://mlsanity.com/paper/2608.30237/) (2026)
- [LaST-R1: Reinforcing Robotic Manipulation via Adaptive Physical Latent Reasoning](https://mlsanity.com/paper/2604.28192/) (2026)
- [Realtime-VLA FLASH: Speculative Inference Framework for Diffusion-based VLAs](https://mlsanity.com/paper/2605.13778/) (2026)
- [IntentVLA: Short-Horizon Intent Modeling for Aliased Robot Manipulation](https://mlsanity.com/paper/2605.14712/) (2026)
- [X-Tokenizer: A Multimodal Action Tokenizer for Vision-Language-Action Pretraining](https://mlsanity.com/paper/2606.14752/) (2026)
- [GEM: Generative Supervision Helps Embodied Intelligence](https://mlsanity.com/paper/2605.28548/) (2026)
- [Data Pyramid for Embodied Manipulation: A Survey](https://mlsanity.com/paper/2607.24744/) (2026)
- [Cloak: Zero-Shot Cross-Embodiment Manipulation by Masking the End-Effector from the VLA](https://mlsanity.com/paper/2606.22836/) (2026)
- [UMI-3D: Extending Universal Manipulation Interface from Vision-Limited to 3D Spatial Perception](https://mlsanity.com/paper/2604.14089/) (2026)
- [HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone](https://mlsanity.com/paper/2607.25895/) (2026)
- [Towards Predictive, Aligned, and Scalable Robot Learning](https://mlsanity.com/paper/2607.11270/) (2026)
- [Knowing When to Stop: Adaptive Action Chunking via Internal Cross-Attention Dynamics in VLAs](https://mlsanity.com/paper/2609.00908/) (2026)
- [VISTA: Vision-Grounded and Physics-Validated Adaptation of UMI data for VLA Training](https://mlsanity.com/paper/2606.04708/) (2026)
- [Cortex 2.0: Grounding World Models in Real-World Industrial Deployment](https://mlsanity.com/paper/2604.20246/) (2026)
- [DyPES-VLA: Learning Shared Dynamics Priors and Embodiment-Specific Control for Cross-Embodiment Manipulation](https://mlsanity.com/paper/2608.06374/) (2026)
- [HybridFlow: A 2-NFE Generative Policy for Real-Time Robotic Manipulation](https://mlsanity.com/paper/2602.13718/) (2026)
- [PACT: Self-Evolving Physical Safety Alignment for Diffusion Policies in Embodied Manipulation](https://mlsanity.com/paper/2606.08414/) (2026)
- [ZETA: A Controlled Study of Zero-Shot Cross-Embodiment VLA Transfer for Tabletop Manipulation](https://mlsanity.com/paper/2609.02546/) (2026)
- [EgoHumanoid-V2: Human-to-Humanoid Transfer of Coordinated Whole-Body Skills for Loco-Manipulation](https://mlsanity.com/paper/2609.37181/) (2026)
- [Encoded but Not in Control: Revealing the Grounding Gap in Vision-Language Robot Policies](https://mlsanity.com/paper/2610.06235/) (2026)
- [RoboAware: Learning to Coordinate Embodied Skills from Counterfactual Outcomes](https://mlsanity.com/paper/2610.11480/) (2026)

## Abstract (arXiv)

Vision-Language-Action (VLA) models hold promise for generalist robotics but currently struggle with data scarcity, architectural inefficiencies, and the inability to generalize across different hardware platforms. We introduce RDT2, a robotic foundation model built upon a 7B parameter VLM designed to enable zero-shot deployment on novel embodiments for open-vocabulary tasks. To achieve this, we collected one of the largest open-source robotic datasets--over 10,000 hours of demonstrations in diverse families--using an enhanced, embodiment-agnostic Universal Manipulation Interface (UMI). Our approach employs a novel three-stage training recipe that aligns discrete linguistic knowledge with continuous control via Residual Vector Quantization (RVQ), flow-matching, and distillation for real-time inference. Consequently, RDT2 becomes one of the first models that simultaneously zero-shot generalizes to unseen objects, scenes, instructions, and even robotic platforms. Besides, it outperforms state-of-the-art baselines in dexterous, long-horizon, and dynamic downstream tasks like playing table tennis. See https://rdt-robotics.github.io/rdt2/ for more information.