Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry

arXiv:2605.20496 · q-bio.NC, cs.CV · Submitted 2026-05-19 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Genomics Radio. Generated commentary on the latest computational biology and genomics papers.

Ines: I'm Ines.

Marcus: And I'm Marcus, here with Yuki.

Ines: We're looking at the paper: "Platonic Representations in the Human Brain".

Marcus: Here's the gist of the paper. The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data.

Ines: This paper investigates whether an analogous geometry can be recovered across human brains using fMRI data.

Marcus: Problem Setup The core problem addressed is testing whether subject-specific fMRI embedding spaces can be translated across subjects using only their intrinsic representational geometry.

Ines: Subjects observe disjoint image sets independently sampled from a distribution D, with no shared images or paired cross-subject correspondences available for learning the translations.

Title and authors: Ines: So this paper is called Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry. It suggests that the way our brains encode things has a universal shape that we can actually recover using just fMRI data from natural scenes, without needing any paired data between subjects.

Marcus: What I’m hearing is they are testing if these subject-specific brain embeddings can be translated across different people using only the intrinsic geometry of those responses, without ever seeing images that belong to both subjects at the same time.

Yuki: It sounds like they are asking if there's a common mathematical structure underlying how different people see and encode natural scenes, and this paper proposes a way to find that structure directly from brain scans.

Ines: Exactly. The core idea is that you can learn embeddings for each subject separately, just by showing them many repeated images, and then use those learned shapes to figure out the relationship between subjects without needing any shared images or training data linking them together initially.

Marcus: They propose a self-supervised encoder that learns these subject-specific fMRI embeddings from repeated stimulus presentations, using voxel reliability weighting and a low-dimensional linear projection with principal component analysis.

Yuki: And then they move on to the next step, which is estimating pairwise brain-to-brain translation estimates called orthogonal rotations, based on geometry derived pseudocorrespondences.

Ines: That’s where things get interesting because it sounds like they are using these geometry-derived pseudocorrespondences to estimate those orthogonal rotations between subjects without having any actual paired cross-subject data for the learning step.

Marcus: They refine those pairwise rotation estimates iteratively, similar to a method called Iterative Closest Point, which is a way to align shapes when you don't have perfect matches.

Yuki: Then they take those pairwise rotations and synchronize them into a single shared latent space by solving an orthogonal synchronization problem over the dimension d, which recovers one rotation per subject mapping all embeddings into a common coordinate system.

Ines: So what they’re doing is taking these independently learned subject spaces and making sure they fit together in one big, shared coordinate system by finding those pairwise rotations for every single person involved.

Marcus: The evaluation was done on the Natural Scenes Dataset, using subject-specific non-shared images to learn the encoders and brain-to-brain translations. They measured performance across two settings: within subjects and across subjects.

Yuki: For within subjects, they found that the embeddings are stable, with an average rank of five point twenty-eight which suggests that even if we look at the same person’s data over time, their representation stays consistent.

Ines: But when they looked across subjects, the synchronization improved retrieval metrics from a score of two point forty-seven to one point ninety-seven and R at one from zero point seventy nine to zero point eighty three. That improvement tells us that aligning these spaces helps them find the right images for each other much better.

Title and authors: Marcus: The results support the existence of an approximately isometric shared neural geometry recoverable directly from fMRI data, which means subject-specific spaces are compatible with a common coordinate system, according to page two of this paper <ref:2605.20496#pg1>.

Yuki: That’s significant because it suggests that we don't need shared stimuli to see if two brains are fundamentally organized in a similar way.

Ines: It has practical implications for cross-subject neural modeling because it means data and encoders learned in one subject can be transferred to another, reducing the need for extensive subject-specific calibration when you move from fMRI to image decoding or vice versa.

Marcus: But they also made a distinction about model brain alignment, suggesting that model features aren't necessarily related by the same near-isometric transformations across subjects.

Yuki: That’s important because it separates the idea of geometric equivalence from what actually happens when you train an AI model on top of those brain representations.

Ines: So, to wrap up this Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry paper, they conclude that subject-specific embeddings learned from brain data alone can be translated across individuals using unsupervised orthogonal rotations without needing paired cross-subject samples.

Marcus: They also found that synchronizing those pairwise rotations into a single shared latent space further indicates that the recovered subject spaces are approximately isometric.

Yuki: That means we can use this method to build a common framework for understanding encoding and pattern component analysis across different subjects, which is what page two of this paper suggests <ref:2605.20496#pg1>.

Ines: It’s a big step because it moves us toward being able to transfer knowledge learned in one person’s brain to another person’s without needing their specific data again.

Marcus: For future work, the authors flag that the method relies on high-quality fMRI data and that the unpaired translation procedure can be sensitive depending on how you initialize those rotations.

Yuki: So while they found this geometry in this setting, they also noted that similar isometries might not hold if we test it in lower signal-to-noise modalities or with stimulus distributions further away from what was used in the Natural Scenes Dataset.

Ines: It’s a necessary caution, because if you apply these findings to other types of data or different visual tasks, that assumed near-isometric relationship might break down.

Marcus: In short, they’ve shown that subject-specific embeddings learned from brain data alone can be translated across individuals using unsupervised orthogonal rotations without paired cross-subject samples.

Yuki: It really suggests a level of universality in the organization of visual processing that is independent of the specific training data used for each individual.

Marcus: For future work, they flag that the method relies on high-quality fMRI data and that the unpaired translation procedure can be sensitive depending on how you initialize those rotations.

Ines: It’s a necessary caution, because if you apply these findings to other types of data or different visual tasks, that assumed near-isometric relationship might break down.

The paper's summary: Ines: So, we're looking at how this paper summarizes its main finding for us. It basically says that even though each person’s brain has its own unique map of what they see, there’s an underlying geometric structure common to all of them.

Marcus: Right. They’re not talking about the actual images themselves, but the mathematical shape of how those images get encoded in the brain scans. They learned subject-specific embeddings from repeated views and then figured out how to translate those across subjects without any paired data between them.

Ines: That’s the core claim: you can map these independent subject spaces into one shared latent space just by figuring out the right orthogonal rotations between them using geometry derived pseudocorrespondences. It means that the distance relationships in one person's brain representation are preserved when you translate it to another person's brain representation.

Marcus: Exactly. The numbers they found were pretty compelling for this part of the math. They showed that within a single subject, the embeddings stay stable, with an average rank of five point two eight, which means their representation doesn't drift much over time.

Ines: But across subjects, when they synchronized those spaces by solving that orthogonal synchronization problem over the dimension d, retrieval metrics got better. They went from two point four seven to one point nine seven for cross-subject retrieval R at one going from zero point seventy nine to zero point eighty three.

Marcus: And that improvement suggests that aligning these spaces actually helps you find the right images much better when you're trying to match brain activity across different people. It supports the idea that subject-specific spaces are compatible with a common coordinate system.

Ines: This has big implications for how we think about neural modeling, Marcus. If you can translate data and encoders learned in one person to another, you don't need to do all that extensive subject-specific calibration every single time you move from fMRI data to image decoding or vice versa.

Marcus: That’s the practical part. It suggests that the underlying structure of visual perception is more universal than we thought, even if the specific pathways differ between individuals. But they did put a stop to it too.

Ines: They made a distinction about model brain alignment, which is important because it suggests that just because neural embeddings are geometrically related doesn't mean the features in an actual AI model are related by those exact same near-isometric transformations across subjects.

Marcus: So, the geometry is there, but it's not a one-to-one mapping for what we see in deep learning models. They’re looking at different things here—the brain structure versus the learned weights of an artificial system.

Ines: And they did flag some limitations too. The method relies on high quality fMRI data, and the way they do that unpaired translation procedure can be sensitive depending on how you initialize those rotations.

Marcus: True. It’s not a plug-and-play solution for any dataset you throw at it without careful tuning of the starting point for those transformations. But overall, they've shown that subject-specific embeddings learned from brain data alone can be translated across individuals using unsupervised orthogonal rotations without needing paired cross-subject samples.

Ines: It really does suggest a level of universality in the organization of visual processing that is independent of the specific training data used for each individual.

The paper's improvements: Tom: So, we’re talking about how the authors suggest they can actually make this method better than what they first proposed. It sounds like they’re looking for ways to refine those rotations and synchronization steps.

Ines: Right. They mention that the method works best when you use high quality fMRI data, which is a big caveat for us, but they also point out that the unpaired translation procedure can be sensitive depending on how you set up those initial rotations.

Marcus: So what they’re saying is that if you want to use this across different datasets or modalities, like maybe lower signal-to-noise data, you need to be extra careful with how you start the process. You can't just throw any initialization at it and expect a good result.

Ines: That makes sense because they’re working with these geometry derived pseudocorrespondences, and if those initial geometric guesses are off by even a little bit, the whole synchronization into that shared latent space might get messy. They need robust methods for getting that starting point right.

Marcus: And they also hint at the future work there. They suggest testing whether this kind of isometry—this near-perfect shared geometry—actually holds up if you move to different stimulus distributions or even lower signal-to-noise modalities than what they used on the Natural Scenes Dataset.

Ines: That’s a big question for the biology side. If this universal geometry we found only works well for natural scenes, does it mean that when we look at human brains responding to something completely different, like a medical scan or a synthetic visual task, this specific kind of shared structure might disappear?

Marcus: And that brings us back to the model brain alignment part. They’ve shown this geometric equivalence in the neural embeddings, but they’re still separating that from what happens when you look at actual AI models. The geometry is one thing, but how an AI learns its weights isn't necessarily governed by these same near-isometric transformations across subjects.

Ines: Exactly. So the implication is that while we can transfer data and encoders between people using this geometric alignment, we shouldn't assume that the underlying functional mapping in a machine learning model will follow those exact same rules when it’s being compared across different subjects or tasks.

Marcus: It means the utility of this method is really focused on cross-subject neural modeling, but it doesn't claim to solve the whole problem of how AI learns representations from scratch. They’re showing a tool for aligning brain data, not a complete theory of general learning transfer.

Ines: So, the takeaway is that we have a way to see if two brains are fundamentally organized in a similar way just by looking at their response geometry, but we have to be cautious about applying that finding outside of natural scenes and remember the difference between neural geometry and artificial model weights.

Conclusion: Ines: So we’re wrapping up this look at "Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry." It really boils down to showing that subject-specific embeddings learned from fMRI data can be translated between individuals using just geometry derived pseudocorrespondences, without needing any paired cross-subject images.

Marcus: Right. The big deal is that this suggests an approximately isometric shared neural geometry is recoverable directly from the fMRI data, meaning the spaces learned for one person are compatible with a common coordinate system across everyone.

Yuki: From a population perspective, it’s interesting because it suggests that even though individual brain responses vary, there's a deep structural similarity in how those responses are organized across different people. It supports the idea that there's some underlying universal organization to visual processing.

Ines: That’s what I was thinking. It means we can actually use data and encoders learned in one subject for another person, which is a massive practical thing if we ever want to do cross-subject neural modeling without needing tons of extra calibration time.

Marcus: It definitely opens up the door for more efficient transfer learning in neuroscience, but remember they also pointed out that model brain alignment isn't necessarily related by those same near-isometric transformations across subjects. That’s a key distinction we have to keep in mind.

Yuki: I just wonder if this universal geometry holds up when we test it on stimulus distributions that are really different from the natural scenes they used for the Natural Scenes Dataset, because that could be a limiting factor for its broader application in population genetics.

Ines: True. That’s the caveat—the method relies on high quality fMRI data and is sensitive to initialization of those rotations, so we can't just assume it works everywhere without careful setup. But the core finding about translating subject spaces remains strong even with those limitations.

Marcus: So, to sum up, they’ve shown that subject-specific embeddings learned from brain data alone can be translated across individuals using unsupervised orthogonal rotations without paired cross-subject samples. It's a way to see if two brains are fundamentally organized in a similar way through their response geometry.

Yuki: I think the real impact here is showing that we don't always need shared stimuli to find universal structures in the brain, which is a big conceptual step for population studies.

Ines: Exactly. This work on "Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry" gives us a new tool to test these geometric hypotheses directly from brain data alone.

Marcus: It’s a powerful technique for aligning subject spaces, but we need to keep an eye on those caveats about initialization and the quality of the fMRI signal when we try to apply it in more complex or diverse settings.

University of Barcelona · Cornell University · Bellvitge Institute for Biomedical Research, Spain

q-bio.NC, cs.CV

Submitted: 2026-05-19

Updated: 2026-11-06

Comments: Code available at https://github.com/memory-formation/platonic-representations-fmri

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 80/100

The gist: The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively…

Key concepts

Subject-Specific Embeddings
These are low-dimensional numerical representations of brain activity learned from a single person's fMRI data based on their unique patterns. The researchers used these to see if they could be understood or mapped onto other people's data without any direct comparison between the subjects.
Orthogonal Rotations (Rs→t)
These are mathematical transformations used to find the specific way one subject's embedding space relates to another subject's. The method uses these rotations, derived from geometry-based pseudocorrespondences, to align the individual spaces into a single shared coordinate system.
Isometric Shared Geometry
This means that the underlying geometric structure of brain representations across different subjects is nearly identical or perfectly preserved. Finding this shared geometry suggests that even though individual brains have unique patterns, they share a fundamental, universal spatial organization.

Terminology

Summary

The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. p. 0

Where the paper says it

Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry Pablo Marcos-Manchón1,2 Rishi Jha3 Lluís Fuentemilla1,2,4 1Department of Cognition, Development and Education Psychology, University of Barcelona 2 Institute of Neurosciences, University of Barcelona 3Department of Computer Science, Cornell University 4Bellvitge Institute for Biomedical Research, Spain Abstract The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains.

This paper investigates whether an analogous geometry can be recovered across human brains using fMRI data. p. 0
Where the paper says it

We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations.

Problem Setup

The core problem addressed is testing whether subject-specific fMRI embedding spaces can be translated across subjects using only their intrinsic representational geometry. Subjects observe disjoint image sets independently sampled from a distribution D, with no shared images or paired cross-subject correspondences available for learning the translations. p. 1

Where the paper says it

Subjects observe disjoint image sets independently sampled from D. No shared images or paired cross-subject correspondences are available for learning the translations.

The goal is to map these independently observed fMRI responses into a shared latent space Z using only their intrinsic representational geometry without paired cross-subject data for learning the mappings. p. 1
Where the paper says it

X(s) ∈ R ns×vs denotes the subject’s neural activity matrix, where rows correspond to image presentations and columns to subject-specific voxel responses. Our goal is to test whether these independently observed fMRI responses can be mapped into a shared latent space Z using only their intrinsic representational geometry, without paired cross-subject data for learning the mappings.

Methodology

The method proceeds in three stages:

  1. A self-supervised encoder learns subject-specific fMRI embeddings from repeated stimulus presentations. p. 0

    Where the paper says it

    We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations.

    This involves voxel reliability weighting to mitigate temporal drifts and a low-dimensional linear projection using PCA. p. 2
    Where the paper says it

    (A) Subject encoder. For each subject, fMRI responses are mapped into a low-dimensional embedding space using voxel reliability weighting, PCA, and multi-view CCA (MCCA), followed by a residual nonlinear refinement trained from repeated stimulus presentations.

  2. Pairwise brain-to-brain translation estimates unsupervised orthogonal rotations Rs→t from geometry-derived pseudocorrespondences. p. 2

    Where the paper says it

    (B) Pairwise brain-to-brain translation. Independently learned subject embeddings are translated between subject pairs by estimating orthogonal rotations Rs→t from geometry-derived pseudocorrespondences.

    This is refined iteratively using a method similar to Iterative Closest Point [26]. p. 3
    Where the paper says it

    These pairs define the initial orthogonal Procrustes problem: R (0) s→t = arg min R∈O(d) ∥Z (s)R − Z˜(t) ∥ 2 F . (8) We refine the translation iteratively using an approach similar to Iterative Closest Point [26].

  3. Pairwise rotations are synchronized into a single shared latent space by solving an orthogonal synchronization problem over O(d). p. 4

    Where the paper says it

    3 Shared latent space construction Given pairwise translations {Rs→t} S s,t=1, we construct a shared latent space Z by solving an orthogonal synchronization problem over O(d) [27, 28]. The goal is to recover one transformation Rs ∈ O(d) per subject such that Rs→t ≈ RsR⊤ t , placing all subjects in a common coordinate system, i.

    This recovers one transformation Rs per subject mapping all embeddings into a common coordinate system. p. 6
    Where the paper says it

    We therefore integrate the pairwise rotations into a all-subject shared latent space by recovering one orthogonal transformation Rs per subject through global synchronization (Subsection 3. 3), mapping all embeddings into a common coordinate system (Fig.

Experimental Evaluation

The method was evaluated on the Natural Scenes Dataset (NSD) [18], using subject-specific, non-shared images to learn encoders and brain-to-brain translations. Performance is measured across two settings: within subjects and across subjects. p. 4

Where the paper says it

We evaluate retrieval in two settings. Within subjects, embeddings from one repetition retrieve the matching image from another repetition among 515 candidates, averaged across repetition pairs.

Within-subject retrieval shows that embeddings are stable, with an average rank of 5. p. 5
Where the paper says it

In general our embeddings are stable: with an average rank of 5. 28, within-subject retrieval works well across subjects.

28. Across subjects, the synchronization improves retrieval metrics from 2. p. 0
Where the paper says it

We show that these independently learned spaces can be translated across subjects using unsupervised orthogonal rotations, without paired cross-subject samples or intermediate model representations. Synchronizing pairwise rotations into a single shared latent space further improves cross-subject retrieval, indicating that subject-specific spaces are mutually compatible with a common coordinate system.

47 to 1.97 and R@1 from 0.79 to 0.83.

Discussion

The results support the existence of an approximately isometric shared neural geometry recoverable directly from fMRI data. This suggests that subject-specific spaces are mutually compatible with a common coordinate system. p. 0

Where the paper says it

We show that these independently learned spaces can be translated across subjects using unsupervised orthogonal rotations, without paired cross-subject samples or intermediate model representations. Synchronizing pairwise rotations into a single shared latent space further improves cross-subject retrieval, indicating that subject-specific spaces are mutually compatible with a common coordinate system.

The findings have practical implications for cross-subject neural modeling by allowing data and encoders learned in one subject to be transferable to another. p. 8
Where the paper says it

This shared neural geometry has practical implications for cross-subject neural modeling. If subjectspecific spaces can be placed into a common coordinate system, then data, encoders, and decoders learned in one subject may become transferable to another, reducing the need for extensive subjectspecific calibration.

Furthermore, the model–brain results separate decodability from geometric equivalence, showing that model features are not related to neural embeddings by the same near-isometric transformations observed across subjects. p. 8
Where the paper says it

Our model–brain results further separate decodability from geometric equivalence. Model features predict neural embeddings under supervised linear mappings, but are not related to them by the same near-isometric transformations observed across brains.

Related Work

The work situates itself within the context of universal geometry in machine learning, which posits that models converge toward a shared latent structure. It contrasts this with existing shared representations in neuroscience, noting that functional alignment methods typically require shared stimuli. p. 0

Where the paper says it

However, existing evidence typically depends on shared stimuli, paired measurements, or external reference spaces. Functional alignment methods for fMRI, such as hyperalignment, map neural responses into common spaces by using shared stimuli to establish cross-subject correspondences [14, 15].

The paper distinguishes its approach by learning subject-specific embeddings from neural repetitions and testing their translation via orthogonal rotations learned from unpaired images. p. 8
Where the paper says it

Other methods introduce anchors through model-derived feature spaces [16] or image-to-fMRI encoders [17], while unsupervised alternatives infer correspondences through structural or distributional matching [45] and optimal transport-based functional alignment [46, 47]. In contrast, we learn subject-specific fMRI embeddings from neural repetitions and test whether they can be translated by orthogonal rotations learned from unpaired subject-specific images.

Model–brain alignment is also examined diagnostically, suggesting that model features are predictive of neural embeddings under supervised linear mappings but not necessarily related by the same near-isometric transformations observed across subjects. p. 8
Where the paper says it

Our model–brain results further separate decodability from geometric equivalence. Model features predict neural embeddings under supervised linear mappings, but are not related to them by the same near-isometric transformations observed across brains.

Conclusion and Limitations

The study concludes that subject-specific embeddings learned from brain data alone can be translated across individuals using unsupervised orthogonal rotations without paired cross-subject samples. The synchronization into a shared latent space further indicates that the recovered subject spaces are approximately isometric. p. 8

Where the paper says it

Synchronizing these rotations into a shared latent space further improved retrieval, indicating that the pairwise translations are mutually compatible with a common coordinate system. Together, these results suggest that the recovered subject spaces are not only similar, but approximately isometric.

Limitations include reliance on high-quality fMRI data and the sensitivity of the unpaired translation procedure to initialization. p. 8
Where the paper says it

Several limitations remain. The analysis relies on high-quality fMRI data with repeated stimulus presentations to isolate stable stimulus-related signal from trial-level noise.

Future work should test whether similar isometries hold in lower-SNR modalities or more distant stimulus distributions. p. 8
Where the paper says it

The analysis relies on high-quality fMRI data with repeated stimulus presentations to isolate stable stimulus-related signal from trial-level noise. It remains unclear whether similar isometries will hold in lower-SNR modalities, smaller datasets, or higher-level cognitive tasks.

The subject-agnostic neural spaces may improve decoding and transfer learning but raise neural privacy concerns if applied without consent. p. 8
Where the paper says it

Future work should test whether shared neural geometries extend to more distant stimulus distributions, task contexts, and clinical datasets. Finally, subject-agnostic neural spaces may improve decoding and transfer learning, but related methods raise neural privacy concerns if applied to sensitive brain data without consent; future use should require explicit consent, careful data governance, and safeguards against misuse.

Acknowledgments

This work was supported by the Spanish Ministerio de Ciencia, Innovación y Universidades, which is part of Agencia Estatal de Investigación (AEI), through the project PID2022 - 140426NB (Co-funded by European Regional Development Fund. ERDF, a way to build Europe). p. 9

Where the paper says it

Acknowledgments and Disclosure of Funding This work was supported by the Spanish Ministerio de Ciencia, Innovación y Universidades, which is part of Agencia Estatal de Investigación (AEI), through the project PID2022 - 140426NB (Co-funded by European Regional Development Fund. ERDF, a way to build Europe).

The authors thank CERCA Programme/Generalitat de Catalunya for institutional support. p. 9
Where the paper says it

ERDF, a way to build Europe). We thank CERCA Programme/Generalitat de Catalunya for institutional support.

R.J. was supported by the Digital Life Initiative Doctoral Fellowship at Cornell Tech. p. 9
Where the paper says it

We thank CERCA Programme/Generalitat de Catalunya for institutional support. was supported by the Digital Life Initiative Doctoral Fellowship at Cornell Tech.

References

[1] Tomas Mikolov, Wen-tau Yih, and Geoffrey Zweig. Linguistic regularities in continuous space word representations. p. 9

Where the paper says it

References [1] Tomas Mikolov, Wen-tau Yih, and Geoffrey Zweig. Linguistic regularities in continuous space word representations.

[2] Maithra Raghu, Justin Gilmer, Jason Yosinski, and Jascha Sohl-Dickstein. p. 9

Where the paper says it

Association for Computational Linguistics, 2013. [2] Maithra Raghu, Justin Gilmer, Jason Yosinski, and Jascha Sohl-Dickstein.

Svcca: singular vector canonical correlation analysis for deep learning dynamics and interpretability. p. 9
Where the paper says it

[2] Maithra Raghu, Justin Gilmer, Jason Yosinski, and Jascha Sohl-Dickstein. Svcca: singular vector canonical correlation analysis for deep learning dynamics and interpretability.

[3] Simon Kornblith, Mohammad Norouzi, Honglak Lee, and Geoffrey Hinton. p. 9

Where the paper says it

ISBN 9781510860964. [3] Simon Kornblith, Mohammad Norouzi, Honglak Lee, and Geoffrey Hinton.

Similarity of neural network representations revisited. p. 9
Where the paper says it

[3] Simon Kornblith, Mohammad Norouzi, Honglak Lee, and Geoffrey Hinton. Similarity of neural network representations revisited.

[4] Ilia Sucholutsky, Lukas Muttenthaler, Adrian Weller, Andi Peng, Andreea Bobu, Been Kim, Bradley C. p. 9

Where the paper says it

In International Conference on Machine Learning (ICML), volume 97, pages 3519–3529, 2019. [4] Ilia Sucholutsky, Lukas Muttenthaler, Adrian Weller, Andi Peng, Andreea Bobu, Been Kim, Bradley C.

Love, Christopher J Cueva, Erin Grant, Iris Groen, Jascha Achterberg, Joshua B. p. 9
Where the paper says it

[4] Ilia Sucholutsky, Lukas Muttenthaler, Adrian Weller, Andi Peng, Andreea Bobu, Been Kim, Bradley C. Love, Christopher J Cueva, Erin Grant, Iris Groen, Jascha Achterberg, Joshua B.

Tenenbaum.

[5] Minyoung Huh, Brian Cheung, Tongzhou Wang, and Phillip Isola. p. 9

Where the paper says it

id=Hiq7lUh4Yn. [5] Minyoung Huh, Brian Cheung, Tongzhou Wang, and Phillip Isola.

The platonic representation hypothesis. p. 0
Where the paper says it

Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry Pablo Marcos-Manchón1,2 Rishi Jha3 Lluís Fuentemilla1,2,4 1Department of Cognition, Development and Education Psychology, University of Barcelona 2 Institute of Neurosciences, University of Barcelona 3Department of Computer Science, Cornell University 4Bellvitge Institute for Biomedical Research, Spain Abstract The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains.

[6] Rishi Dev Jha, Collin Zhang, Vitaly Shmatikov, and John Xavier Morris. p. 9

Where the paper says it

In Proceedings of the 41st International Conference on Machine Learning. [6] Rishi Dev Jha, Collin Zhang, Vitaly Shmatikov, and John Xavier Morris.

Harnessing the universal geometry of embeddings. p. 9
Where the paper says it

[6] Rishi Dev Jha, Collin Zhang, Vitaly Shmatikov, and John Xavier Morris. Harnessing the universal geometry of embeddings.

[7] Guy Dar. p. 9

Where the paper says it

In Advances in Neural Information Processing Systems, 2025. [7] Guy Dar.

mini-vec2vec: Scaling universal geometry alignment with linear transformations. p. 9
Where the paper says it

[7] Guy Dar. mini-vec2vec: Scaling universal geometry alignment with linear transformations, 2026.

[8] Yamini Bansal, Preetum Nakkiran, and Boaz Barak. p. 9

Where the paper says it

org/abs/2510. [8] Yamini Bansal, Preetum Nakkiran, and Boaz Barak.

Revisiting model stitching to compare neural representations. p. 9
Where the paper says it

[8] Yamini Bansal, Preetum Nakkiran, and Boaz Barak. Revisiting model stitching to compare neural representations.

[9] Jianglin Lu, Hailing Wang, Yi Xu, Yizhou Wang, Kuo Yang, and Yun Fu. p. 9

Where the paper says it

id=ak06J5jNR4. [9] Jianglin Lu, Hailing Wang, Yi Xu, Yizhou Wang, Kuo Yang, and Yun Fu.

Representation potentials of foundation models for multimodal alignment: A survey. p. 9
Where the paper says it

[9] Jianglin Lu, Hailing Wang, Yi Xu, Yizhou Wang, Kuo Yang, and Yun Fu. Representation potentials of foundation models for multimodal alignment: A survey.

[10] Uri Hasson, Yuval Nir, Ifat Levy, Galit Fuhrmann, and Rafael Malach. p. 9

Where the paper says it

emnlp-main. [10] Uri Hasson, Yuval Nir, Ifat Levy, Galit Fuhrmann, and Rafael Malach.

Intersubject synchronization of cortical activity during natural vision. p. 9
Where the paper says it

[10] Uri Hasson, Yuval Nir, Ifat Levy, Galit Fuhrmann, and Rafael Malach. Intersubject synchronization of cortical activity during natural vision.

[11] Samuel A Nastase, Valeria Gazzola, Uri Hasson, and Christian Keysers. p. 9

Where the paper says it

1126/science. [11] Samuel A Nastase, Valeria Gazzola, Uri Hasson, and Christian Keysers.

Measuring shared responses across subjects using intersubject correlation. p. 9
Where the paper says it

[11] Samuel A Nastase, Valeria Gazzola, Uri Hasson, and Christian Keysers. Measuring shared responses across subjects using intersubject correlation.

[12] Nikolaus Kriegeskorte, Marieke Mur, and Peter A. p. 9

Where the paper says it

1093/scan/nsz037. [12] Nikolaus Kriegeskorte, Marieke Mur, and Peter A.

Bandettini. p. 9
Where the paper says it

[12] Nikolaus Kriegeskorte, Marieke Mur, and Peter A. Bandettini.

Representational similarity analysis – connecting the branches of systems neuroscience. p. 9
Where the paper says it

Bandettini. Representational similarity analysis – connecting the branches of systems neuroscience.

[13] Baihan Lin and Nikolaus Kriegeskorte. p. 9

Where the paper says it

3389/neuro. [13] Baihan Lin and Nikolaus Kriegeskorte.

The topology and geometry of neural representations. p. 9
Where the paper says it

[13] Baihan Lin and Nikolaus Kriegeskorte. The topology and geometry of neural representations.

[14] James V. Haxby, J. Swaroop Guntupalli, Andrew C. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Connolly, Yaroslav O. Halchenko, Bryan R. Conroy, M. Ida Gobbini, Michael Hanke.

[15] J. Swaroop Guntupalli, Michael Hanke, Yaroslav O. Halchenko, Andrew C. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Connolly, Peter J Ramadge. p. 9
Where the paper says it

1093/scan/nsz037. [12] Nikolaus Kriegeskorte, Marieke Mur, and Peter A.

[16] Qiongyi Zhou, Changde Du, Shengpei Wang, and Huiguang He. p. 9

Where the paper says it

id=Hiq7lUh4Yn. [5] Minyoung Huh, Brian Cheung, Tongzhou Wang, and Phillip Isola.

CLIP-MUSED: CLIP-guided multisubject visual neural information semantic decoding. p. 5
Where the paper says it

21 DINOv2 ViT-S/14 [33] + Ridge Vision Model Guided 24. 46 CLIP ViT-B/16 [34] + Ridge Vision Model Guided 23.

[17] Navve Wasserman, Roman Beliy, Roy Urbach, and Michal Irani. Functional brain-to-brain transformation without shared stimuli. p. 3

Where the paper says it

2 Pairwise brain-to-brain translation To translate embeddings between two subjects, we adapt mini-vec2vec [7] to neural embeddings. Given subjects s and t, we learn an orthogonal transformation Rs→t ∈ O(d) such that Z (s)Rs→t ≈ Z (t) , without paired cross-subject samples during training.

[18] Emily J. Allen, Ghislain St-Yves, Yihan Wu, Jesse L. Breedlove, Jacob S. Prince, Logan T. Dowdle, Matthias Nau.

[19] Thomas T. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Liu. Noise contributions to the fmri signal: An overview. p. 8
Where the paper says it

Several limitations remain. The analysis relies on high-quality fMRI data with repeated stimulus presentations to isolate stable stimulus-related signal from trial-level noise.

[20] Jacob S Prince, Ian Charest, Jan W Kurzawski, John A Pyles, Michael J Tarr, and Kendrick N Kay. p. 9

Where the paper says it

In Proceedings of the 41st International Conference on Machine Learning. [6] Rishi Dev Jha, Collin Zhang, Vitaly Shmatikov, and John Xavier Morris.

Improving the accuracy of single-trial fmri response estimates using glmsingle. p. 1
Where the paper says it

We synchronize pairwise rotations into a single shared latent space, improving cross-subject retrieval and showing that independently learned subject spaces are mutually compatible with a common coordinate system. Our results support the existence of an approximately isometric shared neural geometry recoverable directly from fMRI data, with practical implications for cross-subject neural modeling.

[21] Furkan Ozcelik and Rufin VanRullen. Natural scene reconstruction from fmri signals using generative latent diffusion. p. 0

Where the paper says it

We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations.

[22] Paul Steven Scotti, Atmadeep Banerjee, Jimmie Goode, Stepan Shabalin, Alex Nguyen, Cohen Ethan, Aidan James Dempster.

[23] J. R. KETTENRING. Canonical analysis of several sets of variables. p. 8

Where the paper says it

Several limitations remain. The analysis relies on high-quality fMRI data with repeated stimulus presentations to isolate stable stimulus-related signal from trial-level noise.

[24] Arthur Tenenhaus and Michel Tenenhaus. Regularized generalized canonical correlation analysis. p. 2

Where the paper says it

Let Y ∈ R rns×dPCA denote the concatenation of all views. To extract components shared across repetitions, we apply multi-view canonical correlation analysis (MCCA) to {Yi} r i=1 [23, 24].

[25] Aaron van den Oord, Yazhe Li, and Oriol Vinyals. Representation learning with contrastive predictive coding. p. 9

Where the paper says it

The platonic representation hypothesis. In Proceedings of the 41st International Conference on Machine Learning.

[26] Paul J. Besl and Neil D. McKay. A method for registration of 3-d shapes. p. 1

Where the paper says it

This constraint prevents arbitrary geometric warping and provides a strong test of whether independently learned neural representations can be translated into a shared coordinate system using geometry alone. 3 Method We proceed in three stages.

[27] A Singer. Angular synchronization by eigenvectors and semidefinite programming. p. 4

Where the paper says it

(11) We recover a relaxed solution using a spectral method for orthogonal synchronization [27]. We compute the top-d eigenvectors of B, yielding U ∈ R Sd×d , whose blocks {Us} S s=1 approximate the subject-specific transformations.

[28] Lanhui Wang and Amit Singer. p. 9

Where the paper says it

id=Hiq7lUh4Yn. [5] Minyoung Huh, Brian Cheung, Tongzhou Wang, and Phillip Isola.

Exact and stable recovery of rotations for robust synchronization. p. 5
Where the paper says it

Even for higher-performing subjects, RSA remains around 0. 6, indicating that robust instance-level identification does not require exact preservation of the full pairwise geometry.

[29] Tsung-Yi Lin, Michael Maire, Serge J. p. 3

Where the paper says it

Z¯ lin = 1 r Xr i=1 Y Ui . (3) We then distill these multi-view projections into a single linear mapping via ridge regression: W∗ = arg min W ∥Y W − Z¯ lin∥ 2 F + λreg∥W∥ 2 F , (4) yielding Zlin = Y W∗ ∈ R rns×d .

Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár.

[30] Jacob S. Prince, Ian Charest, Jan W. Kurzawski, John A Pyles. p. 9

Where the paper says it

In Proceedings of the 41st International Conference on Machine Learning. [6] Rishi Dev Jha, Collin Zhang, Vitaly Shmatikov, and John Xavier Morris.

[31] Weiran Wang, Raman Arora, Karen Livescu, and Jeff Bilmes. p. 9

Where the paper says it

id=Hiq7lUh4Yn. [5] Minyoung Huh, Brian Cheung, Tongzhou Wang, and Phillip Isola.

On deep multi-view representation learning. p. 2
Where the paper says it

(A) Subject encoder. For each subject, fMRI responses are mapped into a low-dimensional embedding space using voxel reliability weighting, PCA, and multi-view CCA (MCCA), followed by a residual nonlinear refinement trained from repeated stimulus presentations.

[32] Galen Andrew, Raman Arora, Jeff Bilmes, and Karen Livescu. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Deep canonical correlation analysis. p. 9
Where the paper says it

[2] Maithra Raghu, Justin Gilmer, Jason Yosinski, and Jascha Sohl-Dickstein. Svcca: singular vector canonical correlation analysis for deep learning dynamics and interpretability.

[33] Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy V. Vo, Marc Szafraniec.

[34] Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh. p. 9

Where the paper says it

In International Conference on Machine Learning (ICML), volume 97, pages 3519–3529, 2019. [4] Ilia Sucholutsky, Lukas Muttenthaler, Adrian Weller, Andi Peng, Andreea Bobu, Been Kim, Bradley C.

[35] Andreas Peter Steiner, Alexander Kolesnikov, Xiaohua Zhai, Ross Wightman. p. 9

Where the paper says it

1093/scan/nsz037. [12] Nikolaus Kriegeskorte, Marieke Mur, and Peter A.

[36] Nils Reimers and Iryna Gurevych. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. p. 0

Where the paper says it

Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry Pablo Marcos-Manchón1,2 Rishi Jha3 Lluís Fuentemilla1,2,4 1Department of Cognition, Development and Education Psychology, University of Barcelona 2 Institute of Neurosciences, University of Barcelona 3Department of Computer Science, Cornell University 4Bellvitge Institute for Biomedical Research, Spain Abstract The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains.

[37] Yixuan Li, Jason Yosinski, Jeff Clune, Hod Lipson, and John Hopcroft. p. 9

Where the paper says it

Association for Computational Linguistics, 2013. [2] Maithra Raghu, Justin Gilmer, Jason Yosinski, and Jascha Sohl-Dickstein.

Convergent learning: Do different neural networks learn the same representations. p. 7
Where the paper says it

Learned representations in deep neural networks exhibit structured geometries that reflect statistical properties of the data. Empirical work shows that models trained with different architectures, objectives, and modalities often develop similar representational spaces [1–3, 37].

[38] David Alvarez-Melis and Tommi Jaakkola. Gromov-Wasserstein alignment of word embedding spaces. p. 7

Where the paper says it

Table 4: Supervised model-to-brain alignment. Last-layer model embeddings are mapped to the neural embedding space using each subject’s training images and evaluated on the 515 held-out shared images.

[39] Edouard Grave, Armand Joulin, and Quentin Berthet. Unsupervised alignment of embeddings with wasserstein procrustes. p. 8

Where the paper says it

Other methods introduce anchors through model-derived feature spaces [16] or image-to-fMRI encoders [17], while unsupervised alternatives infer correspondences through structural or distributional matching [45] and optimal transport-based functional alignment [46, 47]. In contrast, we learn subject-specific fMRI embeddings from neural repetitions and test whether they can be translated by orthogonal rotations learned from unpaired subject-specific images.

[40] Luca Moschella, Valentino Maiorca, Marco Fumero, Antonio Norelli, Francesco Locatello.

[41] Guillaume Lample, Alexis Conneau, Marc’Aurelio Ranzato.

[42] Pablo Marcos-Manchón and Lluís Fuentemilla. p. 0

Where the paper says it

Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry Pablo Marcos-Manchón1,2 Rishi Jha3 Lluís Fuentemilla1,2,4 1Department of Cognition, Development and Education Psychology, University of Barcelona 2 Institute of Neurosciences, University of Barcelona 3Department of Computer Science, Cornell University 4Bellvitge Institute for Biomedical Research, Spain Abstract The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains.

Shared representations in brains and models reveal a two-route cortical organization during scene perception. p. 8
Where the paper says it

This could benefit applications such as image-to-fMRI encoding, fMRI-to-image decoding, and synthetic neural data generation, where paired data are expensive and often unavailable across subjects. More broadly, shared spaces provide a route for training neural models from heterogeneous datasets in which different subjects viewed different stimuli.

[43] Janice Chen, Yuan Chang Leong, Christopher J. p. 9

Where the paper says it

[4] Ilia Sucholutsky, Lukas Muttenthaler, Adrian Weller, Andi Peng, Andreea Bobu, Been Kim, Bradley C. Love, Christopher J Cueva, Erin Grant, Iris Groen, Jascha Achterberg, Joshua B.

Honey, Chung H. Yong, Kenneth A. Norman.

[44] Jörn Diedrichsen and Nikolaus Kriegeskorte. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Representational models: A common framework for understanding encoding, pattern-component, and representational-similarity analysis. p. 8
Where the paper says it

These approaches support encoding, decoding, and benchmarking efforts such as Brain-Score [51–56]. Recent work further suggests that brains and models may share lowdimensional representational axes [57].

[45] Daiki Nakamura, Shizuo Kaji, Ryota Kanai, and Ryusuke Hayashi. Unsupervised method for representation transfer from one brain to another. p. 8

Where the paper says it

Our method instead recovers a shared brain space directly from neural data, using model–brain mappings only diagnostically. 6 Discussion The Strong Platonic Representation Hypothesis [6] posits that the universal latent structure of representations not only exists but can be harnessed to translate representations from one space to another without any paired data or model access.

[46] T. Bazeille, H. Richard, H. Janati, and B. Thirion. Local optimal transport for functional brain template estimation. p. 8

Where the paper says it

Functional alignment methods such as hyperalignment map subjects into common spaces, but typically require shared stimuli to establish correspondences [14, 15]. Other methods introduce anchors through model-derived feature spaces [16] or image-to-fMRI encoders [17], while unsupervised alternatives infer correspondences through structural or distributional matching [45] and optimal transport-based functional alignment [46, 47].

[47] Thomas Bazeille, Elizabeth DuPre, Hugo Richard, Jean-Baptiste Poline. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

[48] Daniel L. K. Yamins, Ha Hong, Charles F Cadieu, Ethan A Solomon.

[49] Seyed-Mahdi Khaligh-Razavi and Nikolaus Kriegeskorte. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Deep supervised, but not unsupervised, models may explain IT cortical representation. p. 6
Where the paper says it

These results additionally support approximate isometry across independently-learned subject embeddings: one rotation per subject is sufficient to map all subjects into a common space posited by the Strong Platonic Representation Hypothesis. 4 Model–brain alignment Finally, we test how closely our fMRI embeddings can be mapped to the intermediate representations of artificial neural networks by fitting supervised mappings from our neural embeddings to the final-layer embeddings of four models.

[50] Radoslaw Martin Cichy, Aditya Khosla, Dimitrios Pantazis, Antonio Torralba. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

[51] Martin Schrimpf, Jonas Kubilius, Michael J. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Lee, N. Apurva Ratan Murty.

[52] Charlotte Caucheteux and Jean-Rémi King. Brains and algorithms partially converge in natural language processing. p. 9

Where the paper says it

Representation potentials of foundation models for multimodal alignment: A survey. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 16669–16684.

[53] Talia Konkle and George A. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Alvarez. A self-supervised domain-general learning framework for human ventral stream representation. p. 0
Where the paper says it

We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations.

[54] Aria Y Wang, Kendrick Kay, Thomas Naselaris, Michael J Tarr, Leila Wehbe. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

[55] Adrien Doerig, Tim C. Kietzmann, Emily Allen.

[56] Colin Conwell, Jacob S Prince, Kendrick N Kay, George A Alvarez.

[57] Zirui Chen and Michael F. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Bonner. Universal dimensions of visual representation. p. 8
Where the paper says it

This suggests that current models capture stimulus information relevant to neural responses, while still differing in the geometry of their representational spaces. Brain-derived shared spaces may therefore provide a useful target for identifying which dimensions of representation are shared, distorted, or missing in artificial models.

[58] Matthew F. Glasser, Timothy S Coalson, Emma C Robinson, Carl D Hacker.

[59] Liang Wang, Ryan E. p. 9

Where the paper says it

Dauphin, P. Liang, and J.

B. Mruczek, Michael J Arcaro, and Sabine Kastner. Probabilistic maps of visual topography in human cortex. p. 0
Where the paper says it

Synchronizing pairwise rotations into a single shared latent space further improves cross-subject retrieval, indicating that subject-specific spaces are mutually compatible with a common coordinate system. These results provide evidence for a shared neural geometry in the human visual cortex: subject-specific fMRI representations are approximately isometric across individuals and can be translated through purely geometric transformations.

[60] Ross Wightman. Pytorch image models. p. 8

Where the paper says it

This could benefit applications such as image-to-fMRI encoding, fMRI-to-image decoding, and synthetic neural data generation, where paired data are expensive and often unavailable across subjects. More broadly, shared spaces provide a route for training neural models from heterogeneous datasets in which different subjects viewed different stimuli.

[61] Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. p. 9

Where the paper says it

Tenenbaum, Katherine M. Collins, Katherine Hermann, Kerem Oktar, Klaus Greff, Martin N Hebart, Nathan Cloos, Nikolaus Kriegeskorte, Nori Jacoby, Qiuyi Zhang, Raja Marjieh, Robert Geirhos, Sherol Chen, Simon Kornblith, Sunayana Rane, Talia Konkle, Thomas O’Connell, Thomas Unterthiner, Andrew Kyle Lampinen, Klaus Robert Muller, Mariya Toneva, and Thomas L.

Masked autoencoders are scalable vision learners. p. 0
Where the paper says it

Recent constructive work pushes this hypothesis further: if representations share a common geometry, then embeddings from one model should be translatable into another model’s latent space without shared inputs or paired supervision [6, 7]. However, evidence for this strong form of representational convergence has so far come almost entirely from artificial vision and language systems [5, 8, 9].

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Subject-specific fMRI embeddings can be translated across subjects using only intrinsic geometry via unsupervised orthogonal rotations, which preserve distances and inner products, allowing for cross-subject retrieval on held-out images.

  2. A shared latent space can be constructed by synchronizing pairwise rotations, mapping all subject embeddings into a common coordinate system, which improves cross-subject retrieval and indicates that independently learned subject spaces are mutually compatible with a common coordinate system.

  3. The recovered geometry supports practical applications by suggesting that data, encoders, and decoders learned in one subject may become transferable to another, reducing the need for extensive subject-specific calibration in image-to-fMRI encoding or fMRI-to-image decoding.

Abstract

The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations. We show that these independently learned spaces can be translated across subjects using unsupervised orthogonal rotations, without paired cross-subject samples or intermediate model representations. Synchronizing pairwise rotations into a single shared latent space further improves cross-subject retrieval. These results provide evidence for a shared geometry in learned representations of human visual responses: subject-specific fMRI embeddings can be aligned across individuals within a common coordinate system.

Sources

Related papers