1. Title of the Paper:

How Culturally Aware are Vision-Language Models?

2. Goal of the Paper:

3. Contribution:

4. Methodology:

Dataset Curation:

Model Selection:

연구에서는 다음과 같은 VLMs를 사용

  1. GPT-4V
  2. Gemini Pro Vision
  3. LLaVA
  4. OpenFlamingo