Local Intrinsic Dimension of Representations Predicts Alignment and Generalization in AI Models and Human Brain
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914294712500224 |
|---|---|
| author | Yu, Junjie Ma, Wenxiao Wei, Chen Zhang, Jianyu Deng, Haotian Deng, Zihan Liu, Quanying |
| author_facet | Yu, Junjie Ma, Wenxiao Wei, Chen Zhang, Jianyu Deng, Haotian Deng, Zihan Liu, Quanying |
| contents | Recent work has found that neural networks with stronger generalization tend to exhibit higher representational alignment with one another across architectures and training paradigms. In this work, we show that models with stronger generalization also align more strongly with human neural activity. Moreover, generalization performance, model--model alignment, and model--brain alignment are all significantly correlated with each other. We further show that these relationships can be explained by a single geometric property of learned representations: the local intrinsic dimension of embeddings. Lower local dimension is consistently associated with stronger model--model alignment, stronger model--brain alignment, and better generalization, whereas global dimension measures fail to capture these effects. Finally, we find that increasing model capacity and training data scale systematically reduces local intrinsic dimension, providing a geometric account of the benefits of scaling. Together, our results identify local intrinsic dimension as a unifying descriptor of representational convergence in artificial and biological systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_22722 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Local Intrinsic Dimension of Representations Predicts Alignment and Generalization in AI Models and Human Brain Yu, Junjie Ma, Wenxiao Wei, Chen Zhang, Jianyu Deng, Haotian Deng, Zihan Liu, Quanying Machine Learning Recent work has found that neural networks with stronger generalization tend to exhibit higher representational alignment with one another across architectures and training paradigms. In this work, we show that models with stronger generalization also align more strongly with human neural activity. Moreover, generalization performance, model--model alignment, and model--brain alignment are all significantly correlated with each other. We further show that these relationships can be explained by a single geometric property of learned representations: the local intrinsic dimension of embeddings. Lower local dimension is consistently associated with stronger model--model alignment, stronger model--brain alignment, and better generalization, whereas global dimension measures fail to capture these effects. Finally, we find that increasing model capacity and training data scale systematically reduces local intrinsic dimension, providing a geometric account of the benefits of scaling. Together, our results identify local intrinsic dimension as a unifying descriptor of representational convergence in artificial and biological systems. |
| title | Local Intrinsic Dimension of Representations Predicts Alignment and Generalization in AI Models and Human Brain |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2601.22722 |