Holistic Evaluation of Language Models (HELM), a framework to increase the transparency of language models (https://arxiv.org/abs/2211.09110). This framework is also used to evaluate text-to-image models in Holistic Evaluation of Text-to-Image Models (HEIM) (https://arxiv.org/abs/2311.04287).
Personally I think this has too much clutter and too many buttons - some users won't be able to find their way to the full leaderboard. But I will approval for now.
This is what the landing page for Image2struct looks like (zoomed out for the screenshot here)
The number are not for Image2Struct so don't look at them