mlsanityData to 14 Oct 2026

Shared Geometry Is Not Shared Physics: A Layerwise Test of the Platonic Representation Hypothesis in Astronomy

arXiv 2610.04130 ↗First posted 2 Oct 2026astro-ph.IM cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

In this paper we investigate whether geometrically aligned astronomical representations are also scientifically interchangeable. We analyze every layer's representation from 36 pretrained models using cross-matched optical images, infrared images, and spectra. All 139 analyzed model-survey comparisons show significant local-neighborhood alignment somewhere in the network after accounting for the search over depth. However, alignment does not generally increase with network depth and can be stronger for the changes between consecutive computational blocks than for the block outputs themselves. We find that geometry-selected stages under-perform label-selected stages in every image-transfer comparison; moreover, every final HSC-COSMOS-Web model pair is aligned while every redshift transfer has negative R^2. Models can therefore recover a similar cross-survey geometry among astronomical objects without recovering a survey-invariant linear encoding of the physical properties tested here.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 23 of its 41 references that are in the map. First posted on arXiv 2 Oct 2026.

Builds on

23 of 41 references.

References · 23

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  4. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  5. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  6. 2022A ConvNet for the 2020scited by 3,579
  7. 2019Similarity of Neural Network Representations Revisitedcited by 1,755
  8. 2018On the Spectral Bias of Neural Networkscited by 1,341
  9. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 1,039
  10. 2017A Closer Look at Memorization in Deep Networkscited by 942
  11. 2024Revisiting Feature Prediction for Learning Visual Representations from Videocited by 531
  12. 2024The Platonic Representation Hypothesiscited by 382
11 more
  1. 2024PaliGemma 2: A Family of Versatile VLMs for Transfercited by 207
  2. 2019SGD on Neural Networks Learns Functions of Increasing Complexitycited by 204
  3. 2017Residual Connections Encourage Iterative Inferencecited by 77
  4. 2025Understanding the Emergence of Multimodal Representation Alignmentcited by 31
  5. 2026Back into Plato's Cave: Examining Cross-modal Representational Convergence at Scalecited by 16
  6. 2021Practical Galaxy Morphology Tools from Deep Supervised Representation Learningcited by 15
  7. 2021DeepMerge II: Building Robust Deep Learning Algorithms for Merging Galaxy Identification Across Domainscited by 11
  8. 2025The Platonic Universe: Do Foundation Models See the Same Sky?cited by 6
  9. 2025Toward a Theory of Generalizability in LLM Mechanistic Interpretability Researchcited by 5
  10. 2026What Does a Chemical Language Model Know About Molecules?cited by 2
  11. 2026Not all solutions are created equal: An analytical dissociation of functional and representational similarity in deep linear neural networkscited by 1

Built on by

Not cited yet.