# FocusVLA: Focused Visual Utilization for Vision-Language-Action Models

arXiv: https://arxiv.org/abs/2603.28740 · HTML: https://mlsanity.com/paper/2603.28740/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in VLA architectures and generalization, one of 13 directions in Vision-language-action models.
- Builds on 39 of its 40 references that are in the map.
- First posted on arXiv 30 Mar 2026.

## Placed in

- Vision-language-action models / [VLA architectures and generalization](https://mlsanity.com/vla/vla-architectures-and-generalization/)

## Builds on (most cited first, 30 of 39)

- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics](https://mlsanity.com/paper/2506.01844/) (2025)
- [SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model](https://mlsanity.com/paper/2501.15830/) (2025)
- [RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation](https://mlsanity.com/paper/2506.18088/) (2025)
- [UniVLA: Learning to Act Anywhere with Task-centric Latent Actions](https://mlsanity.com/paper/2505.06111/) (2025)
- [3D-VLA: A 3D Vision-Language-Action Generative World Model](https://mlsanity.com/paper/2403.09631/) (2024)
- [X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model](https://mlsanity.com/paper/2510.10274/) (2025)
- [Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free](https://mlsanity.com/paper/2505.06708/) (2025)
- [RT-H: Action Hierarchies Using Language](https://mlsanity.com/paper/2403.01823/) (2024)
- [VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model](https://mlsanity.com/paper/2509.09372/) (2025)
- [Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Model](https://mlsanity.com/paper/2510.12276/) (2025)
- [RoboCat: A Self-Improving Generalist Agent for Robotic Manipulation](https://mlsanity.com/paper/2306.11706/) (2023)
- [PointVLA: Injecting the 3D World into Vision-Language-Action Models](https://mlsanity.com/paper/2503.07511/) (2025)
- [VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching](https://mlsanity.com/paper/2502.02175/) (2025)
- [InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy](https://mlsanity.com/paper/2510.13778/) (2025)
- [ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver](https://mlsanity.com/paper/2508.10333/) (2025)
- [FlowVLA: Visual Chain of Thought-based Motion Reasoning for Vision-Language-Action Models](https://mlsanity.com/paper/2508.18269/) (2025)

## Built on by (most cited first, 6 of 6)

- [UniTacVLA: Unified Tactile Understanding and Prediction in Vision Language Action Models](https://mlsanity.com/paper/2606.31723/) (2026)
- [Sensing Which Modality Matters: Evidence-Gated Regularization for Robust VLA Policies](https://mlsanity.com/paper/2609.03142/) (2026)
- [An Action Is Worth One Patch: Unified World-Action Modeling with PatchWAM](https://mlsanity.com/paper/2609.25961/) (2026)
- [ActGaze: Learning Action-Grounded Gaze through Counterfactual Visual Interventions for High-Precision Manipulation](https://mlsanity.com/paper/2609.28955/) (2026)
- [Gaze Prompts: Temporally Dense Human Attention for Vision-Language-Action Fine-Tuning](https://mlsanity.com/paper/2609.34550/) (2026)
- [DIVA: Dual-Space Intent-Aware Visual Attenuation for Vision-Language-Action Policies](https://mlsanity.com/paper/2610.09144/) (2026)

## Abstract (arXiv)

Vision-Language-Action (VLA) models improve action generation by conditioning policies on rich vision-language information. However, current auto-regressive policies are constrained by three bottlenecks: (1) architectural bias drives models to overlook visual details, (2) an excessive number of visual tokens makes attention difficult to focus on the correct regions, and (3) task-irrelevant visual information introduces substantial noise - together severely impairing the quality of action. In this paper, we investigate how to effectively utilize different visual representations for action generation. To this end, we first empirically validate the above issues and show that VLA performance is primarily limited by how visual information is utilized, rather than by the quality of visual representations. Based on these insights, we introduce FocusVLA, a novel paradigm that directs the model's attention to task-relevant visual regions to effectively bridge vision to action. Specifically, we first propose Modality Cascaded Attention to eliminate shortcut pathways, thereby compelling VLA models to rely on task-relevant visual details for action generation. Furthermore, we propose Focus Attention, which dynamically selects task-relevant visual patches to control information quantity while explicitly modulating their influence to suppress task-irrelevant noise. Extensive experiments on both simulated and real-world robotic benchmarks demonstrate that FocusVLA not only effectively leverages visual details to perform dexterous manipulations, but also substantially improves performance and accelerates convergence across a variety of tasks.