patrickamadeus/

vlms-are-confused-tourists

Vision Language Models are Confused Tourists ✈️ 🤔

[!NOTE] We are still in the process of beautifying the README of the HF dataset. Nonetheless, our data is fully usable!

Although the cultural dimension has been one of the key aspects in evaluating Vision-Language Models (VLMs), their ability to remain stable across diverse cultural inputs remains largely untested, despite being crucial to support diversity and multicultural societies. Existing evaluations often rely on… See the full description on the dataset page: https://huggingface.co/datasets/patrickamadeus/vlms-are-confused-tourists.