Using Multimodal Deep Neural Networks to Disentangle Language from Visual Aesthetics

• URL: http://arxiv.org/abs/2410.23603v1
• Date: Thu, 31 Oct 2024 03:37:21 GMT
• Title: Using Multimodal Deep Neural Networks to Disentangle Language from Visual Aesthetics
• Authors: Colin Conwell, Christopher Hamblin, Chelsea Boccagno, David Mayo, Jesse Cummings, Leyla Isik, Andrei Barbu,
• Abstract summary: 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.
• Score: 8.749640179057469
• License:
• Abstract: When we experience a visual stimulus as beautiful, how much of that experience derives from perceptual computations we cannot describe versus conceptual knowledge we can readily translate into natural language? Disentangling perception from language in visually-evoked affective and aesthetic experiences through behavioral paradigms or neuroimaging is often empirically intractable. Here, we circumnavigate this challenge by using linear decoding over the learned representations of unimodal vision, unimodal language, and multimodal (language-aligned) 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. Unimodal language models (e.g. GPT2) conditioned on visual embeddings to generate captions (via CLIPCap) yield no further gains. Caption embeddings alone yield less accurate predictions than image and caption embeddings combined (concatenated). 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.

Related papers

• Modelling Multimodal Integration in Human Concept Processing with Vision-and-Language Models [7.511284868070148]
  There is growing evidence that human meaning representations integrate linguistic and sensory-motor information. We investigate whether the integration of multimodal information leads to representations that are more aligned with human brain activity. Our results reveal that VLM representations correlate more strongly than language- and vision-only DNNs with activations in brain areas functionally related to language processing.
  arXiv  Detail & Related papers  (2024-07-25T10:08:37Z)
• MindGPT: Interpreting What You See with Non-invasive Brain Recordings [24.63828455553959]
  We introduce a non-invasive neural decoder, termed as MindGPT, which interprets perceived visual stimuli into natural languages from fMRI signals. Our experiments show that the generated word sequences truthfully represented the visual information conveyed in the seen stimuli.
  arXiv  Detail & Related papers  (2023-09-27T15:35:20Z)
• Unified Language-Vision Pretraining in LLM with Dynamic Discrete Visual Tokenization [52.935150075484074]
  We introduce a well-designed visual tokenizer to translate the non-linguistic image into a sequence of discrete tokens like a foreign language. The resulting visual tokens encompass high-level semantics worthy of a word and also support dynamic sequence length varying from the image. This unification empowers LaVIT to serve as an impressive generalist interface to understand and generate multi-modal content simultaneously.
  arXiv  Detail & Related papers  (2023-09-09T03:01:38Z)
• 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)
• Multimodal Neurons in Pretrained Text-Only Transformers [52.20828443544296]
  We identify "multimodal neurons" that convert visual representations into corresponding text. We show that multimodal neurons operate on specific visual concepts across inputs, and have a systematic causal effect on image captioning.
  arXiv  Detail & Related papers  (2023-08-03T05:27:12Z)
• Seeing in Words: Learning to Classify through Language Bottlenecks [59.97827889540685]
  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.
  arXiv  Detail & Related papers  (2023-06-29T00:24:42Z)
• 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)
• Imagination-Augmented Natural Language Understanding [71.51687221130925]
  We introduce an Imagination-Augmented Cross-modal (iACE) to solve natural language understanding tasks. iACE enables visual imagination with external knowledge transferred from the powerful generative and pre-trained vision-and-language models. Experiments on GLUE and SWAG show that iACE achieves consistent improvement over visually-supervised pre-trained models.
  arXiv  Detail & Related papers  (2022-04-18T19:39:36Z)

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.