Conjuring Semantic Similarity

• URL: http://arxiv.org/abs/2410.16431v1
• Date: Mon, 21 Oct 2024 18:51:34 GMT
• Title: Conjuring Semantic Similarity
• Authors: Tian Yu Liu, Stefano Soatto,
• Abstract summary: The semantic similarity between two textual expressions measures the distance between their latent'meaning' We propose a novel approach whereby the semantic similarity among textual expressions is based not on other expressions they can be rephrased as, but rather based on the imagery they evoke. Our method contributes a novel perspective on semantic similarity that not only aligns with human-annotated scores, but also opens up new avenues for the evaluation of text-conditioned generative models.
• Score: 59.18714889874088
• License:
• Abstract: The semantic similarity between sample expressions measures the distance between their latent 'meaning'. Such meanings are themselves typically represented by textual expressions, often insufficient to differentiate concepts at fine granularity. We propose a novel approach whereby the semantic similarity among textual expressions is based not on other expressions they can be rephrased as, but rather based on the imagery they evoke. While this is not possible with humans, generative models allow us to easily visualize and compare generated images, or their distribution, evoked by a textual prompt. Therefore, we characterize the semantic similarity between two textual expressions simply as the distance between image distributions they induce, or 'conjure.' We show that by choosing the Jensen-Shannon divergence between the reverse-time diffusion stochastic differential equations (SDEs) induced by each textual expression, this can be directly computed via Monte-Carlo sampling. Our method contributes a novel perspective on semantic similarity that not only aligns with human-annotated scores, but also opens up new avenues for the evaluation of text-conditioned generative models while offering better interpretability of their learnt representations.

Related papers

• SemFlow: Binding Semantic Segmentation and Image Synthesis via Rectified Flow [94.90853153808987]
  Semantic segmentation and semantic image synthesis are representative tasks in visual perception and generation. We propose a unified framework (SemFlow) and model them as a pair of reverse problems. Experiments show that our SemFlow achieves competitive results on semantic segmentation and semantic image synthesis tasks.
  arXiv  Detail & Related papers  (2024-05-30T17:34:40Z)
• Interpretable Measures of Conceptual Similarity by Complexity-Constrained Descriptive Auto-Encoding [112.0878081944858]
  Quantifying the degree of similarity between images is a key copyright issue for image-based machine learning. We seek to define and compute a notion of "conceptual similarity" among images that captures high-level relations. Two highly dissimilar images can be discriminated early in their description, whereas conceptually dissimilar ones will need more detail to be distinguished.
  arXiv  Detail & Related papers  (2024-02-14T03:31:17Z)
• Describing Sets of Images with Textual-PCA [89.46499914148993]
  We seek to semantically describe a set of images, capturing both the attributes of single images and the variations within the set. Our procedure is analogous to Principle Component Analysis, in which the role of projection vectors is replaced with generated phrases.
  arXiv  Detail & Related papers  (2022-10-21T17:10:49Z)
• Contrastive Corpus Attribution for Explaining Representations [17.07084455770185]
  Most explanation methods explain a scalar model output. Recent works defined a scalar explanation output: a dot product-based similarity in the representation space to the sample being explained. We propose contrastive corpus similarity, a novel and semantically meaningful scalar explanation output.
  arXiv  Detail & Related papers  (2022-09-30T21:59:10Z)
• Lost in Context? On the Sense-wise Variance of Contextualized Word Embeddings [11.475144702935568]
  We quantify how much the contextualized embeddings of each word sense vary across contexts in typical pre-trained models. We find that word representations are position-biased, where the first words in different contexts tend to be more similar.
  arXiv  Detail & Related papers  (2022-08-20T12:27:25Z)
• Patterns of Lexical Ambiguity in Contextualised Language Models [9.747449805791092]
  We introduce an extended, human-annotated dataset of graded word sense similarity and co-predication. Both types of human judgements indicate that the similarity of polysemic interpretations falls in a continuum between identity of meaning and homonymy. Our dataset appears to capture a substantial part of the complexity of lexical ambiguity, and can provide a realistic test bed for contextualised embeddings.
  arXiv  Detail & Related papers  (2021-09-27T13:11:44Z)
• Image Synthesis via Semantic Composition [74.68191130898805]
  We present a novel approach to synthesize realistic images based on their semantic layouts. It hypothesizes that for objects with similar appearance, they share similar representation. Our method establishes dependencies between regions according to their appearance correlation, yielding both spatially variant and associated representations.
  arXiv  Detail & Related papers  (2021-09-15T02:26:07Z)
• Label Distribution Amendment with Emotional Semantic Correlations for Facial Expression Recognition [69.18918567657757]
  We propose a new method that amends the label distribution of each facial image by leveraging correlations among expressions in the semantic space. By comparing semantic and task class-relation graphs of each image, the confidence of its label distribution is evaluated. Experimental results demonstrate the proposed method is more effective than compared state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-07-23T07:46:14Z)

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.