S$^2$-VLA: State-Space Guided Vision-Language-Action Models for Long-Horizon Manipulation

June 26, 2026 ยท Grace Period ยท ๐Ÿ› IJCAI 2026

โณ Grace Period
This paper is less than 90 days old. We give authors time to release their code before passing judgment.
Authors Zhipeng Xie, Zongyi Han, Xiangyi Wei, Shiliang Sun, Yang Li, Jing Zhao arXiv ID 2606.27872 Category cs.RO: Robotics Cross-listed cs.AI Citations 0 Venue IJCAI 2026
Abstract
Vision-Language-Action (VLA) models have demonstrated strong capabilities in robotic manipulation, but their performance degrades significantly in long-horizon tasks due to cumulative error propagation. This limitation largely arises from static feature fusion mechanisms that rely on fixed weights to combine visual, language, and action representations, preventing the model from adapting to different phases of task execution. To address this limitation, we propose S$^2$-VLA, a framework that introduces a State-Space Guided Adaptive Attention (SSGAA) mechanism. SSGAA maintains a belief state that tracks task progression and generates dynamic gating weights to adaptively fuse information from three complementary sources visual features for spatial perception, task intents for high-level task planning, and temporal action sequences for execution consistency. This adaptive fusion allows the model to shift its focus throughout task execution, aligning with the evolving requirements of different task stages. Despite its compact 2B parameter size, S$^2$-VLA consistently outperforms larger 7B-scale models and achieves state-of-the-art performance on long-horizon manipulation benchmarks, including LIBERO and SimplerEnv. highlighting the importance of adaptive feature fusion for long-horizon robotic manipulation.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Robotics