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 …

While image-to-text models have demonstrated significant advancements in various vision-
language tasks, they remain susceptible to adversarial attacks. Existing white-box attacks on …

Image captioning (IC), bringing vision to language, has drawn extensive attention. A crucial
aspect of IC is the accurate depiction of visual relations among image objects. Visual …

Image captioning is a promising task that attracted researchers in the last few years. Existing
image captioning models are primarily trained to generate one caption per image. However …

Existing mainstream image-text matching methods usually measure the relevance of image-
text pairs by capturing and aggregating the affinities between textual words and visual …

Video captioning aims to automatically generate natural language sentences describing the
content of a video. Although encoder-decoder-based models have achieved promising …

Video captioning aims to automatically describe a video clip with informative sentences. At
present, deep learning-based models have become the mainstream for this task and …

Many studies demonstrate that supervised learning techniques are vulnerable to adversarial
examples. However, adversarial threats in unsupervised learning have not drawn sufficient …

Scene Graph Generation (SGG) plays a pivotal role in downstream vision-language tasks.
Existing SGG methods typically suffer from poor compositional generalizations on unseen …

Recent days have seen significant improvements in multi-modal learning made by Vision-
Language Pre-training (VLP) models. However, most of them employ the coarse-grained …