SUGARCREPE++ Dataset: Vision-Language Model Sensitivity to Semantic and Lexical Alterations

• URL: http://arxiv.org/abs/2406.11171v2
• Date: Wed, 19 Jun 2024 00:03:42 GMT
• Title: SUGARCREPE++ Dataset: Vision-Language Model Sensitivity to Semantic and Lexical Alterations
• Authors: Sri Harsha Dumpala, Aman Jaiswal, Chandramouli Sastry, Evangelos Milios, Sageev Oore, Hassan Sajjad,
• Abstract summary: We introduce the SUGARCREPE++ dataset to analyze the sensitivity of vision-and-language models to semantic alterations. We show that all the models which achieve better performance on compositionality datasets need not perform equally well on SUGARCREPE++.
• Score: 13.608653575298183
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite their remarkable successes, state-of-the-art large language models (LLMs), including vision-and-language models (VLMs) and unimodal language models (ULMs), fail to understand precise semantics. For example, semantically equivalent sentences expressed using different lexical compositions elicit diverging representations. The degree of this divergence and its impact on encoded semantics is not very well understood. In this paper, we introduce the SUGARCREPE++ dataset to analyze the sensitivity of VLMs and ULMs to lexical and semantic alterations. Each sample in SUGARCREPE++ dataset consists of an image and a corresponding triplet of captions: a pair of semantically equivalent but lexically different positive captions and one hard negative caption. This poses a 3-way semantic (in)equivalence problem to the language models. We comprehensively evaluate VLMs and ULMs that differ in architecture, pre-training objectives and datasets to benchmark the performance of SUGARCREPE++ dataset. Experimental results highlight the difficulties of VLMs in distinguishing between lexical and semantic variations, particularly in object attributes and spatial relations. Although VLMs with larger pre-training datasets, model sizes, and multiple pre-training objectives achieve better performance on SUGARCREPE++, there is a significant opportunity for improvement. We show that all the models which achieve better performance on compositionality datasets need not perform equally well on SUGARCREPE++, signifying that compositionality alone may not be sufficient for understanding semantic and lexical alterations. Given the importance of the property that the SUGARCREPE++ dataset targets, it serves as a new challenge to the vision-and-language community.

Related papers

• Towards Semantic Equivalence of Tokenization in Multimodal LLM [149.11720372278273]
  Vision tokenization is essential for semantic alignment between vision and language. This paper proposes a novel dynamic Semantic-Equivalent Vision Tokenizer (SeTok) SeTok groups visual features into semantic units via a dynamic clustering algorithm. The resulting vision tokens effectively preserve semantic integrity and capture both low-frequency and high-frequency visual features.
  arXiv  Detail & Related papers  (2024-06-07T17:55:43Z)
• Q-GroundCAM: Quantifying Grounding in Vision Language Models via GradCAM [3.2688425993442696]
  Many probing studies have revealed that even the best-performing Vision and Language Models (VLMs) struggle to capture aspects of compositional scene understanding. Recent VLM advancements include scaling up both model and dataset sizes, additional training objectives and levels of supervision. This paper introduces a novel suite of quantitative metrics that utilize GradCAM activations to rigorously evaluate the grounding capabilities of pre-trained VLMs.
  arXiv  Detail & Related papers  (2024-04-29T22:06:17Z)
• VISLA Benchmark: Evaluating Embedding Sensitivity to Semantic and Lexical Alterations [13.608653575298183]
  This paper introduces the VISLA benchmark, designed to evaluate the semantic and lexical understanding of language models. An evaluation involving 34 vision-language models (VLMs) and 20 unimodal language models (ULMs) reveals surprising difficulties in distinguishing between lexical and semantic variations.
  arXiv  Detail & Related papers  (2024-04-25T07:08:00Z)
• UniDiff: Advancing Vision-Language Models with Generative and Discriminative Learning [86.91893533388628]
  This paper presents UniDiff, a unified multi-modal model that integrates image-text contrastive learning (ITC), text-conditioned image synthesis learning (IS), and reciprocal semantic consistency modeling (RSC) UniDiff demonstrates versatility in both multi-modal understanding and generative tasks.
  arXiv  Detail & Related papers  (2023-06-01T15:39:38Z)
• Visualizing Linguistic Diversity of Text Datasets Synthesized by Large Language Models [9.808214545408541]
  LinguisticLens is a novel inter-active visualization tool for making sense of and analyzing syntactic diversity of datasets. It supports hierarchical visualization of a text dataset, allowing users to quickly scan for an overview and inspect individual examples.
  arXiv  Detail & Related papers  (2023-05-19T00:53:45Z)
• LANDMARK: Language-guided Representation Enhancement Framework for Scene Graph Generation [34.40862385518366]
  Scene graph generation (SGG) is a sophisticated task that suffers from both complex visual features and dataset longtail problem. We propose LANDMARK (LANguage-guiDed representationenhanceMent frAmewoRK) that learns predicate-relevant representations from language-vision interactive patterns. This framework is model-agnostic and consistently improves performance on existing SGG models.
  arXiv  Detail & Related papers  (2023-03-02T09:03:11Z)
• 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)
• An Empirical Investigation of Commonsense Self-Supervision with Knowledge Graphs [67.23285413610243]
  Self-supervision based on the information extracted from large knowledge graphs has been shown to improve the generalization of language models. We study the effect of knowledge sampling strategies and sizes that can be used to generate synthetic data for adapting language models.
  arXiv  Detail & Related papers  (2022-05-21T19:49:04Z)
• Integrating Language Guidance into Vision-based Deep Metric Learning [78.18860829585182]
  We propose to learn metric spaces which encode semantic similarities as embedding space. These spaces should be transferable to classes beyond those seen during training. This causes learned embedding spaces to encode incomplete semantic context and misrepresent the semantic relation between classes.
  arXiv  Detail & Related papers  (2022-03-16T11:06:50Z)
• Meta-Learning with Variational Semantic Memory for Word Sense Disambiguation [56.830395467247016]
  We propose a model of semantic memory for WSD in a meta-learning setting. Our model is based on hierarchical variational inference and incorporates an adaptive memory update rule via a hypernetwork. We show our model advances the state of the art in few-shot WSD, supports effective learning in extremely data scarce scenarios.
  arXiv  Detail & Related papers  (2021-06-05T20:40:01Z)
• Semantic Complexity in End-to-End Spoken Language Understanding [20.184305170102082]
  We analyze the relationship between the performance of STI models and the difficulty of the use case to which they are applied. We show that near-perfect performance metrics for STI models reported in the literature were obtained with datasets with low semantic complexity values.
  arXiv  Detail & Related papers  (2020-08-06T20:18:53Z)

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.