Automatically determining whether a text and a corresponding image are semantically aligned is a significant challenge for vision-language models, with applications in generative text-to-image and image-to-text tasks. In this work, we study methods for automatic text-image alignment evaluation. We first introduce SeeTRUE: a comprehensive evaluation set, spanning multiple datasets from both text-to-image and image-to-text generation tasks, with human judgements for whether a given text-image pair is semantically aligned. We then describe two automatic methods to determine alignment: the first involving a pipeline based on question generation and visual question answering models, and the second employing an end-to-end classification approach by finetuning multimodal pretrained models. Both methods surpass prior approaches in various text-image alignment tasks, with significant improvements in challenging cases that involve complex composition or unnatural images. Finally, we demonstrate how our approaches can localize specific misalignments between an image and a given text, and how they can be used to automatically re-rank candidates in text-to-image generation.

翻译：自动判断文本与对应图像在语义上是否对齐，是视觉语言模型面临的重要挑战，并在生成式文生图和图生文任务中具有广泛应用。本研究探索了自动评估图文对齐的方法。首先，我们引入SeeTRUE：一个综合评估集，涵盖来自文生图和图生文生成任务的多个数据集，并包含人工标注的语义对齐判断结果。随后，我们描述了两种自动确定对齐的方法：第一种基于问题生成和视觉问答模型的流水线方法，第二种通过微调多模态预训练模型实现端到端分类。两种方法在多种图文对齐任务中均超越先前方法，尤其在处理复杂构图或非自然图像等挑战性案例时表现显著提升。最后，我们展示了这些方法如何定位图像与给定文本之间的具体不一致，以及如何将其用于自动重排文生图生成中的候选结果。