# One-Policy-Fits-All: Geometry-Aware Action Latents for Cross-Embodiment Manipulation

arXiv: https://arxiv.org/abs/2603.14522 · HTML: https://mlsanity.com/paper/2603.14522/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Mar 2026.

## Built on by (most cited first, 5 of 5)

- [Show-Harness: Just a VLM Agent Can Play Robots](https://mlsanity.com/paper/2609.10522/) (2026)
- [Cloak: Zero-Shot Cross-Embodiment Manipulation by Masking the End-Effector from the VLA](https://mlsanity.com/paper/2606.22836/) (2026)
- [ZETA: A Controlled Study of Zero-Shot Cross-Embodiment VLA Transfer for Tabletop Manipulation](https://mlsanity.com/paper/2609.02546/) (2026)
- [GALA: Geometry-Aware Latent Action Modeling for Vision-Language-Action Model Pretraining across Embodiments](https://mlsanity.com/paper/2609.21948/) (2026)
- [SkelWAM: A Skeleton-Guided World-Action Model for Zero-Shot Cross-Embodiment Manipulation](https://mlsanity.com/paper/2609.21983/) (2026)

## Abstract (arXiv)

Cross-embodiment manipulation is crucial for enhancing the scalability of robot manipulation and reducing the high cost of data collection. However, the significant differences between embodiments, such as variations in action spaces and structural disparities, pose challenges for joint training across multiple sources of data. To address this, we propose One-Policy-Fits-All (OPFA), a framework that enables learning a single, versatile policy across multiple embodiments. We first learn a Geometry-Aware Latent Representation (GaLR), which leverages 3D convolution networks and transformers to build a shared latent action space across different embodiments. Then we design a unified latent retargeting decoder that extracts embodiment-specific actions from the latent representations, without any embodiment-specific decoder tuning. OPFA enables end-to-end co-training of data from diverse embodiments, including various grippers and dexterous hands with arbitrary degrees of freedom, significantly improving data efficiency and reducing the cost of skill transfer. We conduct extensive experiments across 11 different end-effectors. The results demonstrate that OPFA significantly improves policy performance in diverse settings by leveraging heterogeneous embodiment data. For instance, cross-embodiment co-training can improve success rates by more than 50% compared to single-source training. Moreover, by adding only a few demonstrations from a new embodiment (e.g., eight), OPFA can achieve performance comparable to that of a well-trained model with 72 demonstrations.