Our abstract paper entitled “Generate Answer to Visual Questions with Pre-trained Vision-and-Language Embeddings” was accepted to WiNLP 2022(paper).