Shengcao1006/MMHal-Bench download history

Shengcao1006/MMHal-Bench is a visual question answering dataset on the Hugging Face Hub. In the last 30 days it was downloaded 1,358 times (249 in the last 7 days), and 30,859 times in total. It ranks #15,360 among datasets by monthly downloads.

MMHal-Bench is a new evaluation benchmark specifically designed for hallucintation in Large Multimodal Models (LMM). It contains 96 challenging questions based on images from OpenImages, and their corresponding ground-truth answers and image contents.

Open Shengcao1006/MMHal-Bench on Hugging Face

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