styfeng/VisCTG

Code for Retrieve, Caption, Generate: Visual Grounding for Enhancing Commonsense in Text Generation Models (https://arxiv.org/abs/2109.03892) AAAI 2022.

VisCTG

Code for Retrieve, Caption, Generate: Visual Grounding for Enhancing Commonsense in Text Generation Models published at AAAI 2022.

Poster and other resources can be found here.

Note: inquiries should be directed to stevenyfeng@gmail.com or by opening an issue here.

Code will be released soon