Large Vision Models Can Solve Mental Rotation Problems
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917229485883392 |
|---|---|
| author | Mason, Sebastian Ray Gjølbye, Anders Højbjerg, Phillip Chavarria Tětková, Lenka Hansen, Lars Kai |
| author_facet | Mason, Sebastian Ray Gjølbye, Anders Højbjerg, Phillip Chavarria Tětková, Lenka Hansen, Lars Kai |
| contents | Mental rotation is a key test of spatial reasoning in humans and has been central to understanding how perception supports cognition. Despite the success of modern vision transformers, it is still unclear how well these models develop similar abilities. In this work, we present a systematic evaluation of ViT, CLIP, DINOv2, and DINOv3 across a range of mental-rotation tasks, from simple block structures similar to those used by Shepard and Metzler to study human cognition, to more complex block figures, three types of text, and photo-realistic objects. By probing model representations layer by layer, we examine where and how these networks succeed. We find that i) self-supervised ViTs capture geometric structure better than supervised ViTs; ii) intermediate layers perform better than final layers; iii) task difficulty increases with rotation complexity and occlusion, mirroring human reaction times and suggesting similar constraints in embedding space representations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_15271 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Large Vision Models Can Solve Mental Rotation Problems Mason, Sebastian Ray Gjølbye, Anders Højbjerg, Phillip Chavarria Tětková, Lenka Hansen, Lars Kai Computer Vision and Pattern Recognition Artificial Intelligence Mental rotation is a key test of spatial reasoning in humans and has been central to understanding how perception supports cognition. Despite the success of modern vision transformers, it is still unclear how well these models develop similar abilities. In this work, we present a systematic evaluation of ViT, CLIP, DINOv2, and DINOv3 across a range of mental-rotation tasks, from simple block structures similar to those used by Shepard and Metzler to study human cognition, to more complex block figures, three types of text, and photo-realistic objects. By probing model representations layer by layer, we examine where and how these networks succeed. We find that i) self-supervised ViTs capture geometric structure better than supervised ViTs; ii) intermediate layers perform better than final layers; iii) task difficulty increases with rotation complexity and occlusion, mirroring human reaction times and suggesting similar constraints in embedding space representations. |
| title | Large Vision Models Can Solve Mental Rotation Problems |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2509.15271 |