mair-lab/vismin download history

mair-lab/vismin is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 723 times (89 in the last 7 days), and 9,411 times in total. It ranks #24,733 among datasets by monthly downloads.

VisMin Dataset Training set: VisMin has 65K samples for training vision-language models (e.g., CLIP, MLLM) for minimal-change understanding. The training set is publicly available. Follow the instructions below to download. Download the Training Set To download the VisMin train

Open mair-lab/vismin on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.