ViStruct: Visual Structural Knowledge Extraction via Curriculum Guided Code-Vision Representation.

Yangyi Chen, Xingyao Wang, Manling Li, Derek Hoiem, Heng Ji

Browse the full EMNLP paper archive.