# UniGraspTransformer: Simplified Policy Distillation for Scalable Dexterous Robotic Grasping

arXiv: https://arxiv.org/abs/2412.02699 · HTML: https://mlsanity.com/paper/2412.02699/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Dec 2024.

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

- [OpenHOI: Open-World Hand-Object Interaction Synthesis with Multimodal Large Language Model](https://mlsanity.com/paper/2505.18947/) (2025)
- [ORV: 4D Occupancy-centric Robot Video Generation](https://mlsanity.com/paper/2506.03079/) (2025)
- [MachaGrasp: Morphology-Aware Cross-Embodiment Dexterous Hand Articulation Generation for Grasping](https://mlsanity.com/paper/2510.06068/) (2025)
- [Contact Map Transfer with Conditional Diffusion Model for Generalizable Dexterous Grasp Generation](https://mlsanity.com/paper/2511.01276/) (2025)
- [Dexterous Grasping with Real-World Robotic Reinforcement Learning](https://mlsanity.com/paper/2503.04014/) (2025)
- [FT-NCFM: An Influence-Aware Data Distillation Framework for Efficient VLA Models](https://mlsanity.com/paper/2511.16233/) (2025)
- [MobileManiBench: Simplifying Model Verification for Mobile Manipulation](https://mlsanity.com/paper/2602.05233/) (2026)
- [THOM: Generating Physically Plausible Hand-Object Meshes From Text](https://mlsanity.com/paper/2604.02736/) (2026)
- [SECOND-Grasp: Semantic Contact-guided Dexterous Grasping](https://mlsanity.com/paper/2605.13117/) (2026)
- [MANGO-Grasp: Mahalanobis Fields over Geometry-Oriented 3D Gaussians for Cross-Embodiment Dexterous Grasping](https://mlsanity.com/paper/2608.02014/) (2026)

## Abstract (arXiv)

We introduce UniGraspTransformer, a universal Transformer-based network for dexterous robotic grasping that simplifies training while enhancing scalability and performance. Unlike prior methods such as UniDexGrasp++, which require complex, multi-step training pipelines, UniGraspTransformer follows a streamlined process: first, dedicated policy networks are trained for individual objects using reinforcement learning to generate successful grasp trajectories; then, these trajectories are distilled into a single, universal network. Our approach enables UniGraspTransformer to scale effectively, incorporating up to 12 self-attention blocks for handling thousands of objects with diverse poses. Additionally, it generalizes well to both idealized and real-world inputs, evaluated in state-based and vision-based settings. Notably, UniGraspTransformer generates a broader range of grasping poses for objects in various shapes and orientations, resulting in more diverse grasp strategies. Experimental results demonstrate significant improvements over state-of-the-art, UniDexGrasp++, across various object categories, achieving success rate gains of 3.5%, 7.7%, and 10.1% on seen objects, unseen objects within seen categories, and completely unseen objects, respectively, in the vision-based setting. Project page: https://dexhand.github.io/UniGraspTransformer.