See, Plan, Rewind: Progress-Aware Vision-Language-Action Models for Robust Robotic Manipulation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911735542185984 |
|---|---|
| author | Dai, Tingjun Han, Mingfei Du, Tingwen Liu, Zhiheng Zhang, Zihao Li, Zhihui Khan, Salman Yu, Jun Chang, Xiaojun |
| author_facet | Dai, Tingjun Han, Mingfei Du, Tingwen Liu, Zhiheng Zhang, Zihao Li, Zhihui Khan, Salman Yu, Jun Chang, Xiaojun |
| contents | Measurement of task progress through explicit, actionable milestones is critical for robust robotic manipulation. This progress awareness enables a model to ground its current task status, anticipate verifiable intermediate states, and detect and recover from failures when progress stalls. To embody this capability, we introduce \textbf{S}ee, \textbf{P}lan, \textbf{R}ewind (SPR), a progress-aware vision-language-action framework that dynamically grounds language instructions into a sequence of spatial subgoals. SPR operates through a continuous core cycle, Seeing the current state and upcoming milestone, Planning a trajectory towards the next 2D waypoint, and Rewinding to a recoverable state upon failure by monitoring progress against the expected sequence. This closed-loop approach enables robust error correction without requiring additional training data or auxiliary models. Extensive experiments demonstrate the framework's effectiveness, generalization and robustness: SPR outperforms the MolmoAct baseline by 5\% on the LIBERO benchmark. On the challenging LIBERO-Plus benchmark with unseen instructions and initial states, SPR achieves state-of-the-art robustness with the smallest performance drop, surpassing OpenVLA-OFT and UniVLA, demonstrating superior out-of-distribution robustness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_09292 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | See, Plan, Rewind: Progress-Aware Vision-Language-Action Models for Robust Robotic Manipulation Dai, Tingjun Han, Mingfei Du, Tingwen Liu, Zhiheng Zhang, Zihao Li, Zhihui Khan, Salman Yu, Jun Chang, Xiaojun Robotics Computer Vision and Pattern Recognition Measurement of task progress through explicit, actionable milestones is critical for robust robotic manipulation. This progress awareness enables a model to ground its current task status, anticipate verifiable intermediate states, and detect and recover from failures when progress stalls. To embody this capability, we introduce \textbf{S}ee, \textbf{P}lan, \textbf{R}ewind (SPR), a progress-aware vision-language-action framework that dynamically grounds language instructions into a sequence of spatial subgoals. SPR operates through a continuous core cycle, Seeing the current state and upcoming milestone, Planning a trajectory towards the next 2D waypoint, and Rewinding to a recoverable state upon failure by monitoring progress against the expected sequence. This closed-loop approach enables robust error correction without requiring additional training data or auxiliary models. Extensive experiments demonstrate the framework's effectiveness, generalization and robustness: SPR outperforms the MolmoAct baseline by 5\% on the LIBERO benchmark. On the challenging LIBERO-Plus benchmark with unseen instructions and initial states, SPR achieves state-of-the-art robustness with the smallest performance drop, surpassing OpenVLA-OFT and UniVLA, demonstrating superior out-of-distribution robustness. |
| title | See, Plan, Rewind: Progress-Aware Vision-Language-Action Models for Robust Robotic Manipulation |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.09292 |