Analyzing Hierarchical Structure in Vision Models with Sparse Autoencoders

• URL: http://arxiv.org/abs/2505.15970v1
• Date: Wed, 21 May 2025 19:38:48 GMT
• Title: Analyzing Hierarchical Structure in Vision Models with Sparse Autoencoders
• Authors: Matthew Lyle Olson, Musashi Hinck, Neale Ratzlaff, Changbai Li, Phillip Howard, Vasudev Lal, Shao-Yen Tseng,
• Abstract summary: The ImageNet hierarchy provides a structured taxonomy of object categories, offering a valuable lens through which to analyze the representations learned by deep vision models.<n>In this work, we conduct a comprehensive analysis of how vision models encode the ImageNet hierarchy, leveraging Sparse Autoencoders (SAEs) to probe their internal representations.
• Score: 6.7161402871287645
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The ImageNet hierarchy provides a structured taxonomy of object categories, offering a valuable lens through which to analyze the representations learned by deep vision models. In this work, we conduct a comprehensive analysis of how vision models encode the ImageNet hierarchy, leveraging Sparse Autoencoders (SAEs) to probe their internal representations. SAEs have been widely used as an explanation tool for large language models (LLMs), where they enable the discovery of semantically meaningful features. Here, we extend their use to vision models to investigate whether learned representations align with the ontological structure defined by the ImageNet taxonomy. Our results show that SAEs uncover hierarchical relationships in model activations, revealing an implicit encoding of taxonomic structure. We analyze the consistency of these representations across different layers of the popular vision foundation model DINOv2 and provide insights into how deep vision models internalize hierarchical category information by increasing information in the class token through each layer. Our study establishes a framework for systematic hierarchical analysis of vision model representations and highlights the potential of SAEs as a tool for probing semantic structure in deep networks.

Related papers

• Language Model as Visual Explainer [72.88137795439407]
  We present a systematic approach for interpreting vision models using a tree-structured linguistic explanation.<n>Our method provides human-understandable explanations in the form of attribute-laden trees.<n>To access the effectiveness of our approach, we introduce new benchmarks and conduct rigorous evaluations.
  arXiv  Detail & Related papers  (2024-12-08T20:46:23Z)
• Emergent Visual-Semantic Hierarchies in Image-Text Representations [13.300199242824934]
  We study the knowledge of existing foundation models, finding that they exhibit emergent understanding of visual-semantic hierarchies. We propose the Radial Embedding (RE) framework for probing and optimizing hierarchical understanding.
  arXiv  Detail & Related papers  (2024-07-11T14:09:42Z)
• Emergent Language Symbolic Autoencoder (ELSA) with Weak Supervision to Model Hierarchical Brain Networks [0.12075823996747355]
  Brain networks display a hierarchical organization, a complexity that poses a challenge for existing deep learning models. We propose a symbolic autoencoder informed by weak supervision and an Emergent Language (EL) framework. Our innovation includes a generalized hierarchical loss function designed to ensure that both sentences and images accurately reflect the hierarchical structure of functional brain networks.
  arXiv  Detail & Related papers  (2024-04-15T13:51:05Z)
• Neural Clustering based Visual Representation Learning [61.72646814537163]
  Clustering is one of the most classic approaches in machine learning and data analysis. We propose feature extraction with clustering (FEC), which views feature extraction as a process of selecting representatives from data. FEC alternates between grouping pixels into individual clusters to abstract representatives and updating the deep features of pixels with current representatives.
  arXiv  Detail & Related papers  (2024-03-26T06:04:50Z)
• Structure-CLIP: Towards Scene Graph Knowledge to Enhance Multi-modal Structured Representations [70.41385310930846]
  We present an end-to-end framework Structure-CLIP to enhance multi-modal structured representations. We use scene graphs to guide the construction of semantic negative examples, which results in an increased emphasis on learning structured representations. A Knowledge-Enhance (KEE) is proposed to leverage SGK as input to further enhance structured representations.
  arXiv  Detail & Related papers  (2023-05-06T03:57:05Z)
• Experimental Observations of the Topology of Convolutional Neural Network Activations [2.4235626091331737]
  Topological data analysis provides compact, noise-robust representations of complex structures. Deep neural networks (DNNs) learn millions of parameters associated with a series of transformations defined by the model architecture. In this paper, we apply cutting edge techniques from TDA with the goal of gaining insight into the interpretability of convolutional neural networks used for image classification.
  arXiv  Detail & Related papers  (2022-12-01T02:05:44Z)
• Perceptual Grouping in Contrastive Vision-Language Models [59.1542019031645]
  We show how vision-language models are able to understand where objects reside within an image and group together visually related parts of the imagery. We propose a minimal set of modifications that results in models that uniquely learn both semantic and spatial information.
  arXiv  Detail & Related papers  (2022-10-18T17:01:35Z)
• Understanding the Role of Individual Units in a Deep Neural Network [85.23117441162772]
  We present an analytic framework to systematically identify hidden units within image classification and image generation networks. First, we analyze a convolutional neural network (CNN) trained on scene classification and discover units that match a diverse set of object concepts. Second, we use a similar analytic method to analyze a generative adversarial network (GAN) model trained to generate scenes.
  arXiv  Detail & Related papers  (2020-09-10T17:59:10Z)
• Look-into-Object: Self-supervised Structure Modeling for Object Recognition [71.68524003173219]
  We propose to "look into object" (explicitly yet intrinsically model the object structure) through incorporating self-supervisions. We show the recognition backbone can be substantially enhanced for more robust representation learning. Our approach achieves large performance gain on a number of benchmarks, including generic object recognition (ImageNet) and fine-grained object recognition tasks (CUB, Cars, Aircraft)
  arXiv  Detail & Related papers  (2020-03-31T12:22:51Z)
• Neural encoding and interpretation for high-level visual cortices based on fMRI using image caption features [14.038605815510145]
  This study introduces one higher-level vision task: image caption (IC) task and proposed the visual encoding model based on IC features (ICFVEM) to encode voxels of high-level visual cortices.
  arXiv  Detail & Related papers  (2020-03-26T08:47:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.