VideoVLA: Video Generators Can Be Generalizable Robot Manipulators
This addresses the problem of deploying robots in open-world environments by exploring a paradigm shift in robot learning through visual imagination.
The paper tackles the problem of limited generalization in robot manipulation by proposing VideoVLA, which transforms video generation models into robotic manipulators that predict both action sequences and future visual outcomes. Experiments show this approach enables strong generalization, including imitating other embodiments' skills and handling novel objects.
Generalization in robot manipulation is essential for deploying robots in open-world environments and advancing toward artificial general intelligence. While recent Vision-Language-Action (VLA) models leverage large pre-trained understanding models for perception and instruction following, their ability to generalize to novel tasks, objects, and settings remains limited. In this work, we present VideoVLA, a simple approach that explores the potential of transforming large video generation models into robotic VLA manipulators. Given a language instruction and an image, VideoVLA predicts an action sequence as well as the future visual outcomes. Built on a multi-modal Diffusion Transformer, VideoVLA jointly models video, language, and action modalities, using pre-trained video generative models for joint visual and action forecasting. Our experiments show that high-quality imagined futures correlate with reliable action predictions and task success, highlighting the importance of visual imagination in manipulation. VideoVLA demonstrates strong generalization, including imitating other embodiments' skills and handling novel objects. This dual-prediction strategy - forecasting both actions and their visual consequences - explores a paradigm shift in robot learning and unlocks generalization capabilities in manipulation systems.