Low-Rank HOCA: Efficient High-Order Cross-Modal Attention for Video Captioning.

Tao Jin, Siyu Huang, Yingming Li, Zhongfei Zhang

Browse the full EMNLP paper archive.