vix.ing · top · new · best · stats

WorldVLA: Towards Autoregressive Action World Model

2025/06/26 by Jun Cen, Cen, Jun, Chaohui Yu +21 · 4 voices · 102 citations
Computer Science · #Action (physics) #Action recognition #Autoregressive model #Data Visualization and Analytics #Generalization #Image (mathematics)

paper · pdf · doi:10.48550/arxiv.2506.21539

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2025/06/26 · openalex created_date 2025/10/15 · openalex updated_date 2026/08/05

Abstract

We present WorldVLA, an autoregressive action world model that unifies action and image understanding and generation. Our WorldVLA intergrates Vision-Language-Action (VLA) model and world model in one single framework. The world model predicts future images by leveraging both action and image understanding, with the purpose of learning the underlying physics of the environment to improve action generation. Meanwhile, the action model generates the subsequent actions based on image observations, aiding in visual understanding and in turn helps visual generation of the world model. We demonstrate that WorldVLA outperforms standalone action and world models, highlighting the mutual enhancement between the world model and the action model. In addition, we find that the performance of the action model deteriorates when generating sequences of actions in an autoregressive manner. This phenomenon can be attributed to the model's limited generalization capability for action prediction, leading to the propagation of errors from earlier actions to subsequent ones. To address this issue, we propose an attention mask strategy that selectively masks prior actions during the generation of the current action, which shows significant performance improvement in the action chunk generation task.

Citations

Cited by

Discussions

Related