# Rendering-Free Lookahead for Question-Guided Active Vision

arXiv: https://arxiv.org/abs/2610.11039 · HTML: https://mlsanity.com/paper/2610.11039/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 34 references that are in the map.
- First posted on arXiv 8 Oct 2026.

## Builds on (most cited first, 8 of 8)

- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026)
- [Cosmos 3: Omnimodal World Models for Physical AI](https://mlsanity.com/paper/2606.02800/) (2026)
- [NavWAM: A Navigation World Action Model for Goal-Conditioned Visual Navigation](https://mlsanity.com/paper/2606.13494/) (2026)
- [Toward Ambulatory Vision: Learning Visually-Grounded Active View Selection](https://mlsanity.com/paper/2512.13250/) (2025)
- [GSMem: 3D Gaussian Splatting as Persistent Spatial Memory for Zero-Shot Embodied Exploration and Reasoning](https://mlsanity.com/paper/2603.19137/) (2026)
- [Planning with the Views](https://mlsanity.com/paper/2605.29563/) (2026)
- [WorldMAP: Bootstrapping Vision-Language Navigation Trajectory Prediction with Generative World Models](https://mlsanity.com/paper/2604.07957/) (2026)

## Abstract (arXiv)

Active robot vision requires controlling the camera to reveal task-relevant information that is hidden from the current viewpoint. For example, determining what is inside a box may require raising the camera and looking down into it. For viewpoint-dependent question answering, the challenge is to select camera motions that expose the visual evidence needed to answer the question. Although vision-language models (VLMs) can interpret observed images, selecting such motions requires anticipating the usefulness of unseen views. We quantify this usefulness as answerability, a VLM's estimate that a view suffices to answer the question, and present Rendering-Free Lookahead (RFL), a viewpoint-selection policy that ranks candidate camera motions by predicted future answerability. RFL transfers visual lookahead from deployment to offline training. At training, a privileged teacher renders candidate future views in 3D Gaussian Splatting (3DGS) scenes and uses a frozen VLM to compute one- and two-step answerability targets. Through two-stage distillation, a student learns to predict these action values from the question, recent visual observations, and a candidate camera motion. At deployment, RFL uses these predicted values to select camera motions without rendering future views. On 377 E3VS-Bench test episodes in unseen environments, RFL improves the mean judge score by 43% over a direct-action baseline using the same VLM. These results support learning camera-control policies from privileged visual lookahead for viewpoint-dependent question answering.