Sitemap

Page Not Found

Curriculum Vitae

Publications

Sitemap

Talks

Teaching

Markdown Generator

Show, Don’t Tell: Evaluating Spatial Cognition in Generative Pixels Rather Than LLM Text

Xu Wang^1,*, Kaixiang Yao^1,*, Miao Pan¹, Xiaohe Zhou¹, Xuanyu Liu¹, Wenqi Zhang^1,‡, Xuhong Zhang^1,†‡

Published in arXiv, 2026

ProVisE introduces protocolized visual evaluation, enabling image-generation models to answer spatial tasks in pixel space while preserving the original benchmark metrics.

Recommended citation: Xu Wang, Kaixiang Yao, Miao Pan, Xiaohe Zhou, Xuanyu Liu, Wenqi Zhang, and Xuhong Zhang. (2026). Show, Don't Tell: Evaluating Spatial Cognition in Generative Pixels Rather Than LLM Text. arXiv:2607.21072.
Download Paper

Code

Spatial-Interactor: Learning Spatial Reasoning through Interaction with the Observable Physical World

Kaixiang Yao^1,*, Xu Wang^1,*, Miao Pan¹, Hu Xiyue¹, Weishi Wang², Daniel Dahlmeier², Jintao Chen¹, Yongliang Shen¹, Xuhong Zhang¹, Wenqi Zhang¹

Published in arXiv, 2026

Recommended citation: Kaixiang Yao, Xu Wang, Miao Pan, Hu Xiyue, Weishi Wang, Daniel Dahlmeier, Jintao Chen, Yongliang Shen, Xuhong Zhang, and Wenqi Zhang. (2026). Spatial-Interactor: Learning Spatial Reasoning through Interaction with the Observable Physical World.
Download Paper

Code