Training-free Deep Concept Injection Enables Language Models for Video Question Answering.

Xudong Lin, Manling Li, Richard S. Zemel, Heng Ji, Shih-Fu Chang

Browse the full EMNLP paper archive.