# FoveaVLA: Foveation Actions Replace the Wrist Camera in VLAs

arXiv: https://arxiv.org/abs/2610.20064 · HTML: https://mlsanity.com/paper/2610.20064/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 47 of its 56 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Placed in

- Vision-language-action models

## Builds on (most cited first, 30 of 47)

- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [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)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset](https://mlsanity.com/paper/2403.12945/) (2024)
- [What Matters in Learning from Offline Human Demonstrations for Robot Manipulation](https://mlsanity.com/paper/2108.03298/) (2021)
- [PaliGemma: A versatile 3B VLM for transfer](https://mlsanity.com/paper/2407.07726/) (2024)
- [RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation](https://mlsanity.com/paper/2410.07864/) (2024)
- [V*: Guided Visual Search as a Core Mechanism in Multimodal LLMs](https://mlsanity.com/paper/2312.14135/) (2023)
- [DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning](https://mlsanity.com/paper/2505.14362/) (2025)
- [Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots](https://mlsanity.com/paper/2402.10329/) (2024)
- [Recurrent Models of Visual Attention](https://mlsanity.com/paper/1406.6247/) (2014)
- [Robotic Control via Embodied Chain-of-Thought Reasoning](https://mlsanity.com/paper/2407.08693/) (2024)
- [MolmoAct: Action Reasoning Models that can Reason in Space](https://mlsanity.com/paper/2508.07917/) (2025)
- [EgoMimic: Scaling Imitation Learning via Egocentric Video](https://mlsanity.com/paper/2410.24221/) (2024)
- [RVT-2: Learning Precise Manipulation from Few Demonstrations](https://mlsanity.com/paper/2406.08545/) (2024)
- [Phantom: Training Robots Without Robots Using Only Human Videos](https://mlsanity.com/paper/2503.00779/) (2025)
- [Vision-Based Manipulators Need to Also See from Their Hands](https://mlsanity.com/paper/2203.12677/) (2022)
- [Vision in Action: Learning Active Perception from Human Demonstrations](https://mlsanity.com/paper/2506.15666/) (2025)
- [CoA-VLA: Improving Vision-Language-Action Models via Visual-Textual Chain-of-Affordance](https://mlsanity.com/paper/2412.20451/) (2024)
- [Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matching](https://mlsanity.com/paper/2512.11130/) (2025)
- [Eye, Robot: Learning to Look to Act with a BC-RL Perception-Action Loop](https://mlsanity.com/paper/2506.10968/) (2025)
- [RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies](https://mlsanity.com/paper/2604.09860/) (2026)
- [WristWorld: Generating Wrist-Views via 4D World Models for Robotic Manipulation](https://mlsanity.com/paper/2510.07313/) (2025)
- [Active Vision Might Be All You Need: Exploring Active Vision in Bimanual Robotic Manipulation](https://mlsanity.com/paper/2409.17435/) (2024)
- [ActiveVLA: Injecting Active Perception into Vision-Language-Action Models for Precise 3D Robotic Manipulation](https://mlsanity.com/paper/2601.08325/) (2026)
- [Gaze-based dual resolution deep imitation learning for high-precision dexterous robot manipulation](https://mlsanity.com/paper/2102.01295/) (2021)

## Abstract (arXiv)

We present FoveaVLA, a Vision-Language-Action (VLA) model that replaces the wrist camera with learned foveation actions. Wrist cameras provide close-up detail but restrict the view to the end-effector's surroundings. External cameras see the whole scene, but the policy receives them at low resolution, so the fine detail needed for manipulation is lost. FoveaVLA recovers this detail by predicting, alongside each action chunk, the interaction region to observe at the next policy query. The policy then receives crops of these regions, taken from the original-resolution images, together with the full views. We build FoveaVLA on π0.5 in two configurations, Single and Dual, using one or two external stereo cameras with full and foveal RGB and depth inputs, and post-train both on DROID with RoboInter interaction annotations. Across four real-world task suites, FoveaVLA-Single achieves 70.8% success with one external RGB-D view, and FoveaVLA-Dual achieves 77.1%, versus 52.1% for the wrist-equipped π0.5. Both match the baseline on plain pick-and-place and improve success with distractors, cloth manipulation, and far-away objects.