arXiv Machine Learning August 21, 2026 By Joonas J\"arve, Halil Ibrahim Aysel, Tarun Khajuria, Meelis Kull

Feature Evolution and Migration during Vision Transformer Training

Read the original on arXiv Machine Learning →

arXiv:2608. 20134v1 Announce Type: cross Abstract: We present a novel view on feature evolution in Vision Transformers (ViTs) by visualizing the training process over two dimensions -- network depth (layer) and training time (epochs).

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Computer Vision

1d ago

Decoding the Functional Roles of Register and High-Norm Patch Tokens in Vision Transformers

arXiv:2610.03698v1 Announce Type: new Abstract: Self-supervised Vision Transformers (ViTs), such as DINOv2, learn rich visual representations, but the functions of their internal tokens remain poorly...

By Neel Varma, Andrew Rufail, Dipika Khullar, Vasu Sharma

llms safety

arXiv AI

Jun 9

RAPID: Layer-Wise Redundancy-Aware Pruning and Importance-Driven Token Merging for Efficient ViT

arXiv:2606. 08156v1 Announce Type: cross Abstract: Vision Transformers (ViTs) achieve strong performance but suffer from high computational costs due to quadratic self-attention complexity.

By Kyumin Choi, Ikbeom Jang

llms efficiency

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 AI

4d ago

Geometric Similarity in VLM Low-Level Vision Representations

The paper introduces GeoSim, a four‑level framework for analyzing how vision‑language models (VLMs) represent low‑level vision tasks. It evaluates hidden‑layer representations across 24 tasks and two VLM paradigms—autoregressive models and diffusion transformers—using global similarity, local geometry, sparse feature decomposition, and topological verification. The study uncovers the organizing principles of low‑level visual representations and highlights their limitations in cross‑task and cross‑model agreement, offering an interpretability lens for assessing latent transferability and diagnosing model‑specific issues.

By Shao-Jun Xia, Huixin Zhang, Zhen Lei, Anlan Sun, Yuner Zhang, Xiaoyang Chen

llms diffusion multimodal safety

arXiv Machine Learning

Jun 16

Analyzing Visual Aircraft Representations with Sparse Autoencoders

arXiv:2606. 15468v1 Announce Type: cross Abstract: Vision models can achieve strong performance on classification tasks, but the internal representations supporting their predictions are often difficult to interpret.

By Deepshik Sharma

arXiv Computer Vision

Sep 18

A Smaller Transformer in Your Transformer

The paper introduces Transformer-Within-Transformer (TWT), a post‑hoc technique that merges contiguous redundant layers in Vision Transformers into a single surrogate layer. By doing so, TWT cuts both parameter count and inference compute while maintaining competitive performance on natural image tasks with only half the depth. In histopathology applications, TWT not only matches but sometimes surpasses the baseline model’s performance.

By Dhananjay Tomar, Marius Aasan, Andreas Kleppe, Ad\'in Ram\'irez Rivera

llms