StaMo: Unsupervised Learning of Generalizable Robot Motion from Compact State Representation
This addresses the problem of inefficient world modeling and decision making in robotics by providing a more balanced state representation, though it appears incremental as it builds on existing VLA-based models and pre-trained components.
The paper tackles the challenge of developing expressive yet compact state representations for embodied intelligence by proposing an unsupervised approach that learns a highly compressed two-token representation using a lightweight encoder and a pre-trained Diffusion Transformer decoder. The method improves performance by 14.3% on LIBERO and 30% in real-world task success, with the token difference naturally serving as an effective latent action that enhances policy co-training by 10.4%.
A fundamental challenge in embodied intelligence is developing expressive and compact state representations for efficient world modeling and decision making. However, existing methods often fail to achieve this balance, yielding representations that are either overly redundant or lacking in task-critical information. We propose an unsupervised approach that learns a highly compressed two-token state representation using a lightweight encoder and a pre-trained Diffusion Transformer (DiT) decoder, capitalizing on its strong generative prior. Our representation is efficient, interpretable, and integrates seamlessly into existing VLA-based models, improving performance by 14.3% on LIBERO and 30% in real-world task success with minimal inference overhead. More importantly, we find that the difference between these tokens, obtained via latent interpolation, naturally serves as a highly effective latent action, which can be further decoded into executable robot actions. This emergent capability reveals that our representation captures structured dynamics without explicit supervision. We name our method StaMo for its ability to learn generalizable robotic Motion from compact State representation, which is encoded from static images, challenging the prevalent dependence to learning latent action on complex architectures and video data. The resulting latent actions also enhance policy co-training, outperforming prior methods by 10.4% with improved interpretability. Moreover, our approach scales effectively across diverse data sources, including real-world robot data, simulation, and human egocentric video.