English

MindJourney: Test-Time Scaling with World Models for Spatial Reasoning

Computer Vision and Pattern Recognition 2025-11-04 v2 Artificial Intelligence Robotics

Abstract

Spatial reasoning in 3D space is central to human cognition and indispensable for embodied tasks such as navigation and manipulation. However, state-of-the-art vision-language models (VLMs) struggle frequently with tasks as simple as anticipating how a scene will look after an egocentric motion: they perceive 2D images but lack an internal model of 3D dynamics. We therefore propose MindJourney, a test-time scaling framework that grants a VLM with this missing capability by coupling it to a controllable world model based on video diffusion. The VLM iteratively sketches a concise camera trajectory, while the world model synthesizes the corresponding view at each step. The VLM then reasons over this multi-view evidence gathered during the interactive exploration. Without any fine-tuning, our MindJourney achieves over an average 7.7% performance boost on the representative spatial reasoning benchmark SAT, showing that pairing VLMs with world models for test-time scaling offers a simple, plug-and-play route to robust 3D reasoning. Meanwhile, our method also improves upon the test-time inference VLMs trained through reinforcement learning, which demonstrates the potential of our method that utilizes world models for test-time scaling.

Keywords

Cite

@article{arxiv.2507.12508,
  title  = {MindJourney: Test-Time Scaling with World Models for Spatial Reasoning},
  author = {Yuncong Yang and Jiageng Liu and Zheyuan Zhang and Siyuan Zhou and Reuben Tan and Jianwei Yang and Yilun Du and Chuang Gan},
  journal= {arXiv preprint arXiv:2507.12508},
  year   = {2025}
}

Comments

Project Page: https://umass-embodied-agi.github.io/MindJourney

R2 v1 2026-07-01T04:04:49.067Z