CAST: Cross-modal Alignment Similarity Test for Vision Language Models

• URL: http://arxiv.org/abs/2409.11007v1
• Date: Tue, 17 Sep 2024 09:14:45 GMT
• Title: CAST: Cross-modal Alignment Similarity Test for Vision Language Models
• Authors: Gautier Dagan, Olga Loginova, Anil Batra,
• Abstract summary: Vision Language Models (VLMs) are typically evaluated with Visual Question Answering (VQA) tasks. We propose a Cross-modal Alignment Similarity Test (CAST) to probe VLMs for self-consistency across modalities.
• Score: 1.679718220022688
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision Language Models (VLMs) are typically evaluated with Visual Question Answering (VQA) tasks which assess a model's understanding of scenes. Good VQA performance is taken as evidence that the model will perform well on a broader range of tasks that require both visual and language inputs. However, scene-aware VQA does not fully capture input biases or assess hallucinations caused by a misalignment between modalities. To address this, we propose a Cross-modal Alignment Similarity Test (CAST) to probe VLMs for self-consistency across modalities. This test involves asking the models to identify similarities between two scenes through text-only, image-only, or both and then assess the truthfulness of the similarities they generate. Since there is no ground-truth to compare against, this evaluation does not focus on objective accuracy but rather on whether VLMs are internally consistent in their outputs. We argue that while not all self-consistent models are capable or accurate, all capable VLMs must be self-consistent.

Related papers

• Trust but Verify: Programmatic VLM Evaluation in the Wild [62.14071929143684]
  Programmatic VLM Evaluation (PROVE) is a new benchmarking paradigm for evaluating VLM responses to open-ended queries. We benchmark the helpfulness-truthfulness trade-offs of a range ofVLMs on PROVE, finding that very few are in-fact able to achieve a good balance between the two.
  arXiv  Detail & Related papers  (2024-10-17T01:19:18Z)
• VHELM: A Holistic Evaluation of Vision Language Models [75.88987277686914]
  We present the Holistic Evaluation of Vision Language Models (VHELM) VHELM aggregates various datasets to cover one or more of the 9 aspects: visual perception, knowledge, reasoning, bias, fairness, multilinguality, robustness, toxicity, and safety. Our framework is designed to be lightweight and automatic so that evaluation runs are cheap and fast.
  arXiv  Detail & Related papers  (2024-10-09T17:46:34Z)
• DARE: Diverse Visual Question Answering with Robustness Evaluation [16.87867803628065]
  Vision Language Models (VLMs) extend remarkable capabilities of text-only large language models and vision-only models. They struggle with a number of crucial vision-language (VL) reasoning abilities such as counting and spatial reasoning. We introduce DARE, Diverse Visual Question Answering with Robustness Evaluation.
  arXiv  Detail & Related papers  (2024-09-26T16:31:50Z)
• VELOCITI: Can Video-Language Models Bind Semantic Concepts through Time? [19.313541287648473]
  VELOCITI is a new benchmark building on complex movie clips to test perception and binding in video language models. Our perception-based tests require discriminating video-caption pairs that share similar entities. Our binding tests require models to associate the correct entity to a given situation while ignoring the different yet plausible entities that also appear in the same video.
  arXiv  Detail & Related papers  (2024-06-16T10:42:21Z)
• Zero-shot Translation of Attention Patterns in VQA Models to Natural Language [65.94419474119162]
  ZS-A2T is a framework that translates the transformer attention of a given model into natural language without requiring any training. We consider this in the context of Visual Question Answering (VQA) Our framework does not require any training and allows the drop-in replacement of different guiding sources.
  arXiv  Detail & Related papers  (2023-11-08T22:18:53Z)
• Can I Trust Your Answer? Visually Grounded Video Question Answering [88.11169242115416]
  We study visually grounded VideoQA in response to the emerging trends of utilizing pretraining techniques for video-language understanding. We construct NExT-GQA -- an extension of NExT-QA with 10.5$K$ temporal grounding labels tied to the original QA pairs.
  arXiv  Detail & Related papers  (2023-09-04T03:06:04Z)
• Counterfactual Samples Synthesizing and Training for Robust Visual Question Answering [59.20766562530209]
  VQA models still tend to capture superficial linguistic correlations in the training set. Recent VQA works introduce an auxiliary question-only model to regularize the training of targeted VQA models. We propose a novel model-agnostic Counterfactual Samples Synthesizing and Training (CSST) strategy.
  arXiv  Detail & Related papers  (2021-10-03T14:31:46Z)
• Counterfactual Samples Synthesizing for Robust Visual Question Answering [104.72828511083519]
  We propose a model-agnostic Counterfactual Samples Synthesizing (CSS) training scheme. CSS generates numerous counterfactual training samples by masking critical objects in images or words in questions. We achieve a record-breaking performance of 58.95% on VQA-CP v2, with 6.5% gains.
  arXiv  Detail & Related papers  (2020-03-14T08:34:31Z)

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.