Davidsonian Scene Graph: Improving Reliability in Fine-grained Evaluation for Text-to-Image Generation

• URL: http://arxiv.org/abs/2310.18235v4
• Date: Wed, 13 Mar 2024 21:58:59 GMT
• Title: Davidsonian Scene Graph: Improving Reliability in Fine-grained Evaluation for Text-to-Image Generation
• Authors: Jaemin Cho, Yushi Hu, Roopal Garg, Peter Anderson, Ranjay Krishna, Jason Baldridge, Mohit Bansal, Jordi Pont-Tuset, Su Wang,
• Abstract summary: We develop an evaluation framework inspired by formal semantics for evaluating text-to-image models. We show that Davidsonian Scene Graph (DSG) produces atomic and unique questions organized in dependency graphs. We also present DSG-1k, an open-sourced evaluation benchmark that includes 1,060 prompts.
• Score: 64.64849950642619
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Evaluating text-to-image models is notoriously difficult. A strong recent approach for assessing text-image faithfulness is based on QG/A (question generation and answering), which uses pre-trained foundational models to automatically generate a set of questions and answers from the prompt, and output images are scored based on whether these answers extracted with a visual question answering model are consistent with the prompt-based answers. This kind of evaluation is naturally dependent on the quality of the underlying QG and VQA models. We identify and address several reliability challenges in existing QG/A work: (a) QG questions should respect the prompt (avoiding hallucinations, duplications, and omissions) and (b) VQA answers should be consistent (not asserting that there is no motorcycle in an image while also claiming the motorcycle is blue). We address these issues with Davidsonian Scene Graph (DSG), an empirically grounded evaluation framework inspired by formal semantics, which is adaptable to any QG/A frameworks. DSG produces atomic and unique questions organized in dependency graphs, which (i) ensure appropriate semantic coverage and (ii) sidestep inconsistent answers. With extensive experimentation and human evaluation on a range of model configurations (LLM, VQA, and T2I), we empirically demonstrate that DSG addresses the challenges noted above. Finally, we present DSG-1k, an open-sourced evaluation benchmark that includes 1,060 prompts, covering a wide range of fine-grained semantic categories with a balanced distribution. We release the DSG-1k prompts and the corresponding DSG questions.

Related papers

• Diversity Enhanced Narrative Question Generation for Storybooks [4.043005183192124]
  We introduce a multi-question generation model (mQG) capable of generating multiple, diverse, and answerable questions. To validate the answerability of the generated questions, we employ a SQuAD2.0 fine-tuned question answering model. mQG shows promising results across various evaluation metrics, among strong baselines.
  arXiv  Detail & Related papers  (2023-10-25T08:10:04Z)
• Modeling What-to-ask and How-to-ask for Answer-unaware Conversational Question Generation [30.086071993793823]
  What-to-ask and how-to-ask are the two main challenges in the answer-unaware setting. We present SG-CQG, a two-stage CQG framework.
  arXiv  Detail & Related papers  (2023-05-04T18:06:48Z)
• GrapeQA: GRaph Augmentation and Pruning to Enhance Question-Answering [19.491275771319074]
  Commonsense question-answering (QA) methods combine the power of pre-trained Language Models (LM) with the reasoning provided by Knowledge Graphs (KG) A typical approach collects nodes relevant to the QA pair from a KG to form a Working Graph followed by reasoning using Graph Neural Networks(GNNs) We propose GrapeQA with two simple improvements on the WG: (i) Prominent Entities for Graph Augmentation identifies relevant text chunks from the QA pair and augments the WG with corresponding latent representations from the LM, and (ii) Context-Aware Node Pruning removes nodes that are less relevant to the QA
  arXiv  Detail & Related papers  (2023-03-22T05:35:29Z)
• Toward Unsupervised Realistic Visual Question Answering [70.67698100148414]
  We study the problem of realistic VQA (RVQA), where a model has to reject unanswerable questions (UQs) and answer answerable ones (AQs) We first point out 2 drawbacks in current RVQA research, where (1) datasets contain too many unchallenging UQs and (2) a large number of annotated UQs are required for training. We propose a new testing dataset, RGQA, which combines AQs from an existing VQA dataset with around 29K human-annotated UQs. This combines pseudo UQs obtained by randomly pairing images and questions, with an
  arXiv  Detail & Related papers  (2023-03-09T06:58:29Z)
• Generative Language Models for Paragraph-Level Question Generation [79.31199020420827]
  Powerful generative models have led to recent progress in question generation (QG) It is difficult to measure advances in QG research since there are no standardized resources that allow a uniform comparison among approaches. We introduce QG-Bench, a benchmark for QG that unifies existing question answering datasets by converting them to a standard QG setting.
  arXiv  Detail & Related papers  (2022-10-08T10:24:39Z)
• Unified Question Generation with Continual Lifelong Learning [41.81627903996791]
  Existing QG methods mainly focus on building or training models for specific QG datasets. We propose a model named UnifiedQG based on lifelong learning techniques, which can continually learn QG tasks. In addition, we transform the ability of a single trained Unified-QG model in improving $8$ Question Answering (QA) systems' performance.
  arXiv  Detail & Related papers  (2022-01-24T14:05:18Z)
• Improving Unsupervised Question Answering via Summarization-Informed Question Generation [47.96911338198302]
  Question Generation (QG) is the task of generating a plausible question for a passage, answer> pair. We make use of freely available news summary data, transforming declarative sentences into appropriate questions using dependency parsing, named entity recognition and semantic role labeling. The resulting questions are then combined with the original news articles to train an end-to-end neural QG model.
  arXiv  Detail & Related papers  (2021-09-16T13:08:43Z)
• C3VQG: Category Consistent Cyclic Visual Question Generation [51.339348810676896]
  Visual Question Generation (VQG) is the task of generating natural questions based on an image. In this paper, we try to exploit the different visual cues and concepts in an image to generate questions using a variational autoencoder (VAE) without ground-truth answers. Our approach solves two major shortcomings of existing VQG systems: (i) minimize the level of supervision and (ii) replace generic questions with category relevant generations.
  arXiv  Detail & Related papers  (2020-05-15T20:25:03Z)
• Asking and Answering Questions to Evaluate the Factual Consistency of Summaries [80.65186293015135]
  We propose an automatic evaluation protocol called QAGS (pronounced "kags") to identify factual inconsistencies in a generated summary. QAGS is based on the intuition that if we ask questions about a summary and its source, we will receive similar answers if the summary is factually consistent with the source. We believe QAGS is a promising tool in automatically generating usable and factually consistent text.
  arXiv  Detail & Related papers  (2020-04-08T20:01:09Z)

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.