academiasinicanlplab / multi-vqg Goto Github PK
View Code? Open in Web Editor NEWMVQG collects engaging questions corresponding to image sequences related to human events. This dataset was proposed to EMNLP 2022 paper-- Multi-VQG: Generating Engaging Questions for Multiple Images
License: MIT License