Login / Signup

Q: How to Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A: Self-Train on Unlabeled Images!

Zaid KhanBG Vijay KumarSamuel SchulterXiang YuYun FuManmohan Chandraker
Published in: CVPR (2023)
Keyphrases
  • language model
  • information retrieval
  • computer vision
  • training data
  • n gram
  • clustering algorithm
  • document retrieval
  • similarity measure
  • low level
  • data sources
  • data points
  • co occurrence