Generating Image Descriptions via Sequential Cross-Modal Alignment Guided by Human Gaze.

Ece Takmaz, Sandro Pezzelle, Lisa Beinborn, Raquel Fernndez

Browse the full EMNLP paper archive.