VL2Lite: Task-Specific Knowledge Distillation from Large Vision-Language Models to Lightweight Networks.

Jinseong Jang, Chunfei Ma, Byeongwon Lee

Browse the full CVPR paper archive.