๐Ÿ“–
FocusVLA: Future-Intent Guided Dynamic View Routing for Vision-Language-Action Models
Huixin Xu, Ziya Yu, Peng Li

FocusVLA: Future-Intent Guided Dynamic View Routing for Vision-Language-Action Models

Author / Organization:
Science & Mathematics Published 2026 Verification Pending

Overview & Synopsis

Vision-language-action (VLA) models have become an important paradigm for robotic manipulation by unifying visual perception, language understanding, and action generation. Since reliable manipulation requires sufficient observation of objects, end-effectors, and their spatial relationships, multi-view inputs are widely used to provide complementary spatial information. However, our experiments show that existing VLA models do not fully exploit such multi-view complementarity. The policy tends to persistently rely on the wrist view, lacking the ability to dynamically adjust global-local view dependency across manipulation stages, which limits stage adaptability and robustness. To address this issue, we propose FocusVLA, a stage-aware multi-view routing framework. FocusVLA predicts future interaction intent from historical actions and uses it to guide dynamic view routing during action generation. By adaptively coordinating global and local observations across different task stages, FocusVLA improves multi-view utilization. On RoboTwin 2.0, FocusVLA improves the average success rate over the strongest baseline $ฯ€_{0.5}$ by 10.10 pp. In real-world evaluation, FocusVLA improves the success rate from 71.7\% to 85.0\%, achieving a 13.3\% gain over $ฯ€_{0.5}$. Moreover, the intent prediction and dynamic routing modules of FocusVLA can be integrated into existing VLA frameworks without modifying their core architectures.

Indexed from repository: arXiv (arxiv.org)
Loading document from verified source...
Keyboard Shortcuts: ← / → for pages • + / - for zoom • 0 to fit width Powered by Mozilla PDF.js v3.11

Document Information

Document Title FocusVLA: Future-Intent Guided Dynamic View Routing for Vision-Language-Action Models
Author / Organization Huixin Xu, Ziya Yu, Peng Li
Subject & Discipline cs.RO
Categories Science & Mathematics
Publication Year 2026
License / Rights arXiv Open Access / Creative Commons
Resource Verification Listed PDF (Verification Pending)
Original Host Website arXiv (arxiv.org)
File Format application/pdf (Portable Document Format)
Language EN
Catalog ID DOC-263413
Original Source Link Visit Source Page on arxiv.org »