Robotics: Science and Systems XXII

BagelVLA: Enhancing Long-Horizon Manipulation via Interleaved Vision-Language-Action Generation

Yucheng Hu, Jianke Zhang, Yuanfei Luo, Yanjiang Guo, Xiaoyu Chen, Sun Xinshu, Kun Feng, Qingzhou Lu, Sheng Chen, Yangang Zhang, Wei Li, Jianyu Chen

Abstract:

Equipping embodied agents with the ability to reason about tasks, foresee physical outcomes, and generate precise actions is essential for general-purpose manipulation. While recent Vision-Language-Action (VLA) models have leveraged pre-trained foundation models, they typically focus on either linguistic planning or visual forecasting in isolation. These methods rarely integrate both capabilities simultaneously to guide action generation, leading to suboptimal performance in complex, long-horizon manipulation tasks. To bridge this gap, we propose BagelVLA, a unified model that integrates linguistic planning, visual forecasting, and action generation within a single framework. Initialized from a pretrained unified understanding and generative model, BagelVLA is trained to interleave textual reasoning and visual prediction directly into the action execution loop. To efficiently couple these modalities, we introduce Residual Flow Guidance (RFG), which initializes from current observation and leverages single-step denoising to extract predictive visual features, guiding action generation with minimal latency. Extensive experiments demonstrate that BagelVLA outperforms existing baselines by a significant margin on multiple simulated and real-world benchmarks, particularly in tasks requiring multi-stage reasoning.

Download:

Bibtex:

  
@INPROCEEDINGS{HuY2-RSS-26, 
    AUTHOR    = {Yucheng Hu AND Jianke Zhang AND Yuanfei Luo AND Yanjiang Guo AND Xiaoyu Chen AND Sun Xinshu AND Kun Feng AND Qingzhou Lu AND Sheng Chen AND Yangang Zhang AND Wei Li AND Jianyu Chen}, 
    TITLE     = {{BagelVLA: Enhancing Long-Horizon Manipulation via Interleaved Vision-Language-Action Generation}}, 
    BOOKTITLE = {Proceedings of Robotics: Science and Systems}, 
    YEAR      = {2026}, 
    ADDRESS   = {Sydney, Australia}, 
    MONTH     = {July}, 
    DOI       = {10.15607/RSS.2026.XXII.083} 
}