Seeing in Words: Learning to Classify through Language Bottlenecks

• URL: http://arxiv.org/abs/2307.00028v1
• Date: Thu, 29 Jun 2023 00:24:42 GMT
• Title: Seeing in Words: Learning to Classify through Language Bottlenecks
• Authors: Khalid Saifullah, Yuxin Wen, Jonas Geiping, Micah Goldblum, Tom Goldstein
• Abstract summary: Humans can explain their predictions using succinct and intuitive descriptions. We show that a vision model whose feature representations are text can effectively classify ImageNet images.
• Score: 59.97827889540685
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Neural networks for computer vision extract uninterpretable features despite achieving high accuracy on benchmarks. In contrast, humans can explain their predictions using succinct and intuitive descriptions. To incorporate explainability into neural networks, we train a vision model whose feature representations are text. We show that such a model can effectively classify ImageNet images, and we discuss the challenges we encountered when training it.

Related papers

• Using Multimodal Deep Neural Networks to Disentangle Language from Visual Aesthetics [8.749640179057469]
  We use linear decoding over the learned representations of unimodal vision, unimodal language, and multimodal deep neural network (DNN) models to predict human beauty ratings of naturalistic images. We show that unimodal vision models (e.g. SimCLR) account for the vast majority of explainable variance in these ratings. Language-aligned vision models (e.g. SLIP) yield small gains relative to unimodal vision. Taken together, these results suggest that whatever words we may eventually find to describe our experience of beauty, the ineffable computations of feedforward perception may provide sufficient foundation for that experience.
  arXiv  Detail & Related papers  (2024-10-31T03:37:21Z)
• Representing visual classification as a linear combination of words [0.0]
  We present an explainability strategy that uses a vision-language model to identify language-based descriptors of a visual classification task. By leveraging a pre-trained joint embedding space between images and text, our approach estimates a new classification task as a linear combination of words. We find that the resulting descriptors largely align with clinical knowledge despite a lack of domain-specific language training.
  arXiv  Detail & Related papers  (2023-11-18T02:00:20Z)
• DeViL: Decoding Vision features into Language [53.88202366696955]
  Post-hoc explanation methods have often been criticised for abstracting away the decision-making process of deep neural networks. In this work, we would like to provide natural language descriptions for what different layers of a vision backbone have learned. We train a transformer network to translate individual image features of any vision layer into a prompt that a separate off-the-shelf language model decodes into natural language.
  arXiv  Detail & Related papers  (2023-09-04T13:59:55Z)
• TExplain: Explaining Learned Visual Features via Pre-trained (Frozen) Language Models [14.019349267520541]
  We propose a novel method that leverages the capabilities of language models to interpret the learned features of pre-trained image classifiers. Our approach generates a vast number of sentences to explain the features learned by the classifier for a given image. Our method, for the first time, utilizes these frequent words corresponding to a visual representation to provide insights into the decision-making process.
  arXiv  Detail & Related papers  (2023-09-01T20:59:46Z)
• Localization vs. Semantics: Visual Representations in Unimodal and Multimodal Models [57.08925810659545]
  We conduct a comparative analysis of the visual representations in existing vision-and-language models and vision-only models. Our empirical observations suggest that vision-and-language models are better at label prediction tasks. We hope our study sheds light on the role of language in visual learning, and serves as an empirical guide for various pretrained models.
  arXiv  Detail & Related papers  (2022-12-01T05:00:18Z)
• 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)
• Controlled Caption Generation for Images Through Adversarial Attacks [85.66266989600572]
  We study adversarial examples for vision and language models, which typically adopt a Convolutional Neural Network (i.e., CNN) for image feature extraction and a Recurrent Neural Network (RNN) for caption generation. In particular, we investigate attacks on the visual encoder's hidden layer that is fed to the subsequent recurrent network. We propose a GAN-based algorithm for crafting adversarial examples for neural image captioning that mimics the internal representation of the CNN.
  arXiv  Detail & Related papers  (2021-07-07T07:22:41Z)
• Leveraging Sparse Linear Layers for Debuggable Deep Networks [86.94586860037049]
  We show how fitting sparse linear models over learned deep feature representations can lead to more debuggable neural networks. The resulting sparse explanations can help to identify spurious correlations, explain misclassifications, and diagnose model biases in vision and language tasks.
  arXiv  Detail & Related papers  (2021-05-11T08:15:25Z)
• AttViz: Online exploration of self-attention for transparent neural language modeling [7.574392147428978]
  We propose AttViz, an online toolkit for exploration of self-attention---real values associated with individual text tokens. We show how existing deep learning pipelines can produce outputs suitable for AttViz, offering novel visualizations of the attention heads and their aggregations with minimal effort, online.
  arXiv  Detail & Related papers  (2020-05-12T12:21:40Z)

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.