TensorX
返回文献探索

Paper · arXiv 2606.01247

Where to Look: Can Foundation Models Reach a Target Viewpoint Through Active Exploration?

Liyang Li, Muzhi Zhu, Zhiyue Zhao, Hengyu Zhao, Ke Liu, Linhao Zhong, Hao Chen, Chunhua Shen

32 upvotesMay 31, 2026arXiv 预印本
AI 摘要

Target Viewpoint Reproduction task challenges foundation models to actively adjust 3D viewpoints to match target images, revealing limitations in visual history processing and embodied movement mapping, with a unified post-training framework improving success rates through various training methods.

Target Viewpoint ReproductionTVRBenchembodied AIvisual historyspatial intelligencepost-training frameworkexpert-trajectory SFTrationale-supervised CoT-SFToffline Single-turn GRPOon-policy Multi-turn GRPO

Abstract

Humans can reproduce the viewpoint specified by a target image through active head and body motion, yet spatial intelligence in foundation models has largely been studied as passive understanding of pre-collected observations. We introduce Target Viewpoint Reproduction (TVR) -- an active task where an agent adjusts its viewpoint in a 3D environment until its observation matches a given target image -- and TVRBench, an indoor-simulation benchmark spanning scene scale and target-view visual richness. TVR is far from solved: on the evaluation split, the strongest open-source and closed-source models reach only 7.8% and 12.0% success. Fine-grained analysis identifies two consistent bottlenecks: off-the-shelf models struggle with multi-turn visual history, and performance drops sharply when viewpoint reproduction requires body translation rather than in-place rotation, exposing a gap in mapping spatial discrepancies to embodied movement. To study reducing this gap, we build a unified TVR post-training framework covering expert-trajectory SFT, rationale-supervised CoT-SFT, offline Single-turn GRPO, and on-policy Multi-turn GRPO from live simulator rollouts. Visual-action SFT supplies the main gain, raising a 9B open-source model to 50.8% success; Multi-turn GRPO provides targeted multi-room refinement and reaches 51.4% overall, while CoT supervision and Single-turn GRPO degrade closed-loop performance. These results establish TVRBench as a testbed for measuring and training foundation models that actively perceive and act in 3D environments. Our code, data, and models are available at https://github.com/aim-uofa/TVRBench.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
Where to Look: Can Foundation Models Reach a Target Viewpoint Through Active Exploration? | TensorX