English
Related papers

Related papers: Can Cross-Layer Transcoders Replace Vision Transfo…

200 papers

Transformers are popular neural network models that use layers of self-attention and fully-connected nodes with embedded tokens. Vision Transformers (ViT) adapt transformers for image recognition tasks. In order to do this, the images are…

Computer Vision and Pattern Recognition · Computer Science 2023-04-28 Brian Kenji Iwana , Akihiro Kusuda

A key goal in mechanistic interpretability is circuit analysis: finding sparse subgraphs of models corresponding to specific behaviors or capabilities. However, MLP sublayers make fine-grained circuit analysis on transformer-based language…

Machine Learning · Computer Science 2024-11-08 Jacob Dunefsky , Philippe Chlenski , Neel Nanda

Generative Vision-Language Models (VLMs) perform well on multimodal reasoning, but how visual inputs are transformed to text remains poorly understood. Existing interpretability work on VLMs uses Sparse Autoencoders (SAEs), which decompose…

Machine Learning · Computer Science 2026-05-25 Dimitrios Damianos , Leon Voukoutis , Georgios Skyrianos , Vassilis Katsouros , Georgios Paraskevopoulos

How do vision transformers (ViTs) represent and process the world? This paper addresses this long-standing question through the first systematic analysis of 6.6K features across all layers, extracted via sparse autoencoders, and by…

Computer Vision and Pattern Recognition · Computer Science 2025-09-23 Jinyeong Kim , Junhyeok Kim , Yumin Shim , Joohyeok Kim , Sunyoung Jung , Seong Jae Hwang

Mechanistic interpretability seeks to understand how Large Language Models (LLMs) represent and process information. Recent approaches based on dictionary learning and transcoders enable representing model computation in terms of sparse,…

As deep learning models increasingly find applications in critical domains such as medical imaging, the need for transparent and trustworthy decision-making becomes paramount. Many explainability methods provide insights into how these…

Computer Vision and Pattern Recognition · Computer Science 2023-11-09 Piotr Komorowski , Hubert Baniecki , Przemysław Biecek

Vision Transformers (ViTs) are increasingly utilized in various computer vision tasks due to their powerful representation capabilities. However, it remains understudied how ViTs process information layer by layer. Numerous studies have…

Computer Vision and Pattern Recognition · Computer Science 2025-04-01 Teresa Dorszewski , Lenka Tětková , Robert Jenssen , Lars Kai Hansen , Kristoffer Knutsen Wickstrøm

To truly understand vision models, we must not only interpret their learned features but also validate these interpretations through controlled experiments. While earlier work offers either rich semantics or direct control, few post-hoc…

Computer Vision and Pattern Recognition · Computer Science 2025-11-25 Samuel Stevens , Wei-Lun Chao , Tanya Berger-Wolf , Yu Su

As a special type of transformer, Vision Transformers (ViTs) are used to various computer vision applications (CV), such as image recognition. There are several potential problems with convolutional neural networks (CNNs) that can be solved…

Computer Vision and Pattern Recognition · Computer Science 2022-11-14 Sonain Jamil , Md. Jalil Piran , Oh-Jin Kwon

Vision transformers (ViTs) are quickly becoming the de-facto architecture for computer vision, yet we understand very little about why they work and what they learn. While existing studies visually analyze the mechanisms of convolutional…

Computer Vision and Pattern Recognition · Computer Science 2022-12-14 Amin Ghiasi , Hamid Kazemi , Eitan Borgnia , Steven Reich , Manli Shu , Micah Goldblum , Andrew Gordon Wilson , Tom Goldstein

Vision Transformers (ViTs) have become prominent models for solving various vision tasks. However, the interpretability of ViTs has not kept pace with their promising performance. While there has been a surge of interest in developing {\it…

Computer Vision and Pattern Recognition · Computer Science 2025-05-02 Yao Qiang , Chengyin Li , Prashant Khanduri , Dongxiao Zhu

Mechanistic interpretability seeks to uncover how internal components of neural networks give rise to predictions. A persistent challenge, however, is disentangling two often conflated notions: decodability--the recoverability of…

Machine Learning · Computer Science 2025-10-14 Lianghuan Huang , Yingshan Chang

The emergence of vision transformers (ViTs) in image classification has shifted the methodologies for visual representation learning. In particular, ViTs learn visual representation at full receptive field per layer across all the image…

Computer Vision and Pattern Recognition · Computer Science 2024-08-05 Li Zhang , Jiachen Lu , Sixiao Zheng , Xinxuan Zhao , Xiatian Zhu , Yanwei Fu , Tao Xiang , Jianfeng Feng , Philip H. S. Torr

Though vision transformers (ViTs) have achieved state-of-the-art performance in a variety of settings, they exhibit surprising failures when performing tasks involving visual relations. This begs the question: how do ViTs attempt to perform…

Computer Vision and Pattern Recognition · Computer Science 2024-11-26 Michael A. Lepori , Alexa R. Tartaglini , Wai Keen Vong , Thomas Serre , Brenden M. Lake , Ellie Pavlick

Recent work has explored how individual components of the CLIP-ViT model contribute to the final representation by leveraging the shared image-text representation space of CLIP. These components, such as attention heads and MLPs, have been…

Computer Vision and Pattern Recognition · Computer Science 2024-10-22 Sriram Balasubramanian , Samyadeep Basu , Soheil Feizi

Multilingual Large Language Models (LLMs) can process many languages, yet how they internally represent this diversity remains unclear. Do they form shared multilingual representations with language-specific decoding, and if so, why does…

Computation and Language · Computer Science 2026-02-10 Abir Harrasse , Florent Draye , Punya Syon Pandey , Zhijing Jin , Bernhard Schölkopf

Multilayer perceptrons (MLPs) are an integral part of large language models, yet their dense representations render them difficult to understand, edit, and steer. Recent methods learn interpretable approximations via neuron-level sparsity,…

Machine Learning · Computer Science 2026-01-15 James Oldfield , Shawn Im , Sharon Li , Mihalis A. Nicolaou , Ioannis Patras , Grigorios G Chrysos

Sparse autoencoders (SAEs) extract human-interpretable features from deep neural networks by transforming their activations into a sparse, higher dimensional latent space, and then reconstructing the activations from these latents.…

Machine Learning · Computer Science 2025-02-13 Gonçalo Paulo , Stepan Shabalin , Nora Belrose

Recently vision transformers (ViT) have been applied successfully for various tasks in computer vision. However, important questions such as why they work or how they behave still remain largely unknown. In this paper, we propose an…

Computer Vision and Pattern Recognition · Computer Science 2022-10-19 Van-Anh Nguyen , Khanh Pham Dinh , Long Tung Vuong , Thanh-Toan Do , Quan Hung Tran , Dinh Phung , Trung Le

Vision Transformers (ViTs) are normally regarded as a stack of transformer layers. In this work, we propose a novel view of ViTs showing that they can be seen as ensemble networks containing multiple parallel paths with different lengths.…

Computer Vision and Pattern Recognition · Computer Science 2023-08-15 Shuning Chang , Pichao Wang , Hao Luo , Fan Wang , Mike Zheng Shou
‹ Prev 1 2 3 10 Next ›