open-vision-language.github.io

About Open-Vision-Language

This is a website with an easy-to-remember sub-domain name, to conveniently host scientific projects and results about vision and language.

The current member projects of this domain includes:

  1. SuTI: An in-context subject-driven text-to-image generator that draws subject-specific images without fine-tuning.
  2. OVEN: Datasets and Empirical Results for a novel task of ``Open-domain Visual Entity Recognition’’.
  3. InfoSeek: A new VQA dataset that evaluates multimodal LLMs on answering visual infomation-seeking questions.