arXiv Computer Vision
Decoding the Functional Roles of Register and High-Norm Patch Tokens in Vision Transformers
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Computer Vision
4d ago
Subtoken Vision Transformer for Fine-grained Recognition
arXiv:2607.09086v2 Announce Type: replace Abstract: We present Subtoken Vision Transformer (SubViT), a selective image tokenization method for fine-grained visual recognition. Standard Vision Transfo...
arXiv Machine Learning
Jul 9
Activation Quantization of Vision Encoders Needs Prefixing Registers
arXiv:2510. 04547v5 Announce Type: replace Abstract: Large pretrained vision encoders are central to multimodal intelligence, powering applications from on-device vision processing to vision-language models.
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.
arXiv AI
Aug 12
Token-Based Detection of Spurious Correlations in Vision Transformers
arXiv:2509. 04009v2 Announce Type: replace-cross Abstract: Due to their powerful feature association capabilities, neural network-based computer vision models have the ability to detect and exploit unintended patterns within the data, potentially leading to correct predictions based on incorrect or unintended but statistically relevant signals.
arXiv Computer Vision
Sep 17
Decoder-Agnostic Token Merging for Vision Transformers: A Systematic Study of G2TM
The paper studies Graph-Guided Token Merging (G2TM), a module that reduces token count in Vision Transformers. It evaluates G2TM across multiple segmentation frameworks and decoder types, finding that its performance gains are tied to the encoder rather than the decoder. The authors report consistent reductions in GFLOPs (22‑47%) and throughput improvements (up to 74%) on ADE20K, with optimal hyperparameters depending mainly on backbone pre‑training and target dataset.
arXiv Machine Learning
Aug 21
Feature Evolution and Migration during Vision Transformer Training
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).