arXiv Machine Learning July 8, 2026 By Prashant C. Raju

Geometric Stability: The Missing Axis of Representations

Read the original on arXiv Machine Learning →

arXiv:2601. 09173v5 Announce Type: replace Abstract: Representational similarity analysis and related methods compare the internal geometries of neural networks, but they measure only alignment between spaces, leaving a blind spot -- whether a representation's structure is reliably recoverable, not merely similar.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

safety

arXiv AI

Jun 17

Robustness of Similarity-based Positional Encoding Under Rotations: Theoretical Analysis and Experimental Validation

arXiv:2606. 17961v1 Announce Type: cross Abstract: Positional encoding is a fundamental component of Transformer architectures, as it injects information about the spatial or sequential arrangement of inputs.

By Andrea Santomauro, Luigi Portinale, Giorgio Leonardi

llms computer-vision benchmarks

arXiv Machine Learning

Jul 3

Transformer Geometry Observatory TGO-II: Representational Similarity Observatory

arXiv:2607. 02386v1 Announce Type: cross Abstract: While Vision Transformers have achieved remarkable success across computer vision and language applications, the geometric evolution of their internal representations throughout training remains insufficiently understood.

By Kaustubh Kapil, Kishor P. Upla

llms computer-vision safety

arXiv Machine Learning

Aug 10

Equivariant Sparse Autoencoders: Mechanistic Interpretability of Neural Networks on Symmetric Data

arXiv:2511. 09432v2 Announce Type: replace Abstract: Machine learning (ML) models achieve remarkable performance but remain hard to interpret due to their scale and complexity.

By Ege Erdogan, Ana Lucic

safety

arXiv Machine Learning

Jun 2

Positional Encodings Anchor Spatial Structure in Vision Transformers: A Geometric Perspective on Robustness

arXiv:2606. 00124v1 Announce Type: cross Abstract: Positional embeddings (PEs) in Vision Transformers (ViTs) are known to impact performance and robustness, but their role in shaping internal spatial representations is not well understood.

By Mahmoud Mannes

llms rag

arXiv Machine Learning

Jun 10

Encoding the Euler Characteristic Transform

arXiv:2606. 10824v1 Announce Type: new Abstract: The Euler Characteristic Curve (ECC) records the Euler characteristic of a linearly embedded cell complex as a function of filtration height in a given direction, and the Euler Characteristic Transform (ECT) is the injective shape descriptor obtained by collecting ECCs over many directions.

By Nello Blaser, Odin Hoff Gardaa, Lars M. Salbu, Elena Xinyi Wang, Bastian Rieck

llms nlp benchmarks safety

Hugging Face Trending Papers

Aug 11

Iterative Erasure Count Is Not an Affine-Invariant Concept Dimension

How many directions does a neural representation use to encode a concept? A common answer repeatedly erases probe directions and reports the stopping count or cumulative removed rank.

Related stories

Robustness of Similarity-based Positional Encoding Under Rotations: Theoretical Analysis and Experimental Validation

Transformer Geometry Observatory TGO-II: Representational Similarity Observatory

Equivariant Sparse Autoencoders: Mechanistic Interpretability of Neural Networks on Symmetric Data

Positional Encodings Anchor Spatial Structure in Vision Transformers: A Geometric Perspective on Robustness

Encoding the Euler Characteristic Transform

Iterative Erasure Count Is Not an Affine-Invariant Concept Dimension