Brain-Language Model Alignment: Insights into the Platonic Hypothesis and Intermediate-Layer Advantage
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911222528475136 |
|---|---|
| author | López-Cardona, Ángela Idesis, Sebastián Masias-Bruns, Mireia Abadal, Sergi Arapakis, Ioannis |
| author_facet | López-Cardona, Ángela Idesis, Sebastián Masias-Bruns, Mireia Abadal, Sergi Arapakis, Ioannis |
| contents | Do brains and language models converge toward the same internal representations of the world? Recent years have seen a rise in studies of neural activations and model alignment. In this work, we review 25 fMRI-based studies published between 2023 and 2025 and explicitly confront their findings with two key hypotheses: (i) the Platonic Representation Hypothesis -- that as models scale and improve, they converge to a representation of the real world, and (ii) the Intermediate-Layer Advantage -- that intermediate (mid-depth) layers often encode richer, more generalizable features. Our findings provide converging evidence that models and brains may share abstract representational structures, supporting both hypotheses and motivating further research on brain-model alignment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_17833 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Brain-Language Model Alignment: Insights into the Platonic Hypothesis and Intermediate-Layer Advantage López-Cardona, Ángela Idesis, Sebastián Masias-Bruns, Mireia Abadal, Sergi Arapakis, Ioannis Neurons and Cognition Artificial Intelligence Do brains and language models converge toward the same internal representations of the world? Recent years have seen a rise in studies of neural activations and model alignment. In this work, we review 25 fMRI-based studies published between 2023 and 2025 and explicitly confront their findings with two key hypotheses: (i) the Platonic Representation Hypothesis -- that as models scale and improve, they converge to a representation of the real world, and (ii) the Intermediate-Layer Advantage -- that intermediate (mid-depth) layers often encode richer, more generalizable features. Our findings provide converging evidence that models and brains may share abstract representational structures, supporting both hypotheses and motivating further research on brain-model alignment. |
| title | Brain-Language Model Alignment: Insights into the Platonic Hypothesis and Intermediate-Layer Advantage |
| topic | Neurons and Cognition Artificial Intelligence |
| url | https://arxiv.org/abs/2510.17833 |