This work proposes FOCAL-VLA, a framework that combines subtask-guided geometry distillation with implicit world modeling to learn representations of current spatial structure and future interaction dynamics, and shows that FOCAL-VLA outperforms baselines on both simulation benchmarks and real-world manipulation tasks.
Abstract
Vision-language-action (VLA) models built on pretrained vision-language models have demonstrated strong performance across diverse robotic manipulation tasks. However, VLA models that directly map current 2D observations to actions often lack sufficient spatial and temporal understanding, limiting their performance in precise and long-horizon manipulation. Recent methods enhance VLA models through geometric supervision and future-state prediction across the entire scene. However, these methods can suffer from redundant scene information, distracting the model from learning the geometry and dynamics relevant to the current interaction. To address this issue, we propose FOCAL-VLA, a framework that combines subtask-guided geometry distillation with implicit world modeling to learn representations of current spatial structure and future interaction dynamics. To focus geometric learning on the current subtask, we transfer geometric knowledge from VGGT to the VLA model by aligning geometry latents with features from subtask-relevant image regions. To capture the future 3D evolution of the current interaction, we incorporate implicit world modeling using Track4World features from current and future demonstration frames. The two complementary representations jointly guide action generation without running VGGT or Track4World at inference time. Experiments show that FOCAL-VLA outperforms baselines on both simulation benchmarks and real-world manipulation tasks. Project website: https://zhiyuan-gao.github.io/FOCAL-VLA/.
V-Link is proposed, which explicitly recovers visual representations during the vision-language (VL) to action (A) feature transfer and injects them into Action DiT through asymmetric pathways.
Ye-Hao Lu, Jia-Rui Yang, Yu-Ning Su et al.· 0 citations
Geometry-Change VLA (GC-VLA), which learns to predict multiview future-current geometry-change tokens from current observations and applies Geometry-Conditioned Residual Flow (GCRF), using a binary intervention router and a single bounded residual velocity policy learned from closed-loop feedback.
Jinu Pahk, Jesoon Kang, T. Park et al.· 0 citations
Existing vision-language-action (VLA) models benefit from large-scale vision-language pretraining, but tend to memorize training trajectories, limiting their ability to adapt learned interactions to new task and scene configurations. To address this problem, we propose \textbf{Region-Aware Interaction Networks (RAIN)},...
Seung-Hun Lee, Kyuhwan Shim, Ye-Jun Song et al.· 0 citations
H-VLA is proposed, a hierarchical VLA framework that decouples high-level key-action reasoning from low-level motion generation and adopts a two-stage training strategy that emphasizes key-action reasoning during pre-training and dense motion generation during fine-tuning.
Xiong-Feng Peng, Lu Xu, Yan-Dong Wang et al.· arXiv.org· 0 citations
GaussianWAM is proposed, a training-time representation-enhancement framework that organizes geometric and semantic supervision through a 3D Gaussian field and improves performance on standard LIBERO and shows positive transfer trends on RoboTwin and real-world manipulation.
Zi-Jian Zhang, Yu-Qing Jiang, Wei-Tao Zhou et al.· 3 citations
Vision-language-action (VLA) models provide a unified framework for visual understanding, language grounding, and robot control, yet anticipating how the scene will evolve beyond the current observation can provide additional information for control. Action supervision alone does not explicitly encourage such anticipat...
Zi-Le Zhong, Yu-Xin Du, Tao Lin et al.· 0 citations
From feet to fingertips — we are teaching robots intelligent whole-body control, fine dexterity, and teamwork to complete a broad range of complex tasks.