HL Dataset: Visually-grounded Description of Scenes, Actions and Rationales.

Michele Cafagna, Kees van Deemter, Albert Gatt

Browse the full INLG paper archive.