Image captioning, also called report generation in medical field, aims to describe visual
content of images in human language, which requires to model semantic relationship …

If you ask a human to describe an image, they might do so in a thousand different ways.
Traditionally, image captioning models are trained to generate a single" best"(most like a …

Traditional automated metrics for evaluating conditional natural language generation use
pairwise comparisons between a single generated text and the best-matching gold-standard …

With the introduction of transformer-based models for vision and language tasks, such as
LLaVA and Chameleon, there has been renewed interest in the discrete tokenized …

Text-to-video retrieval has seen significant advancements, yet the ability of models to
discern subtle differences in captions still requires verification. In this paper, we introduce a …

Abstract Language is the predominant means by which humans communicate and
accumulate knowledge acquired through our senses, with vision being the most valued of …

If you ask a human to describe an image, they might do so in a thousand different ways.
Each of these descriptions depends not only on the image but also on a rich tapestry of …