afs07fda89sdfas90/data download history
afs07fda89sdfas90/data is a visual question answering dataset on the Hugging Face Hub. In the last 30 days it was downloaded 591 times (526 in the last 7 days), and 1,821 times in total. It ranks #28,954 among datasets by monthly downloads.
OccludeBench: Can VLMs See the Hint? OccludeBench tests whether vision-language models use the specific visual evidence a question requires. Each of the 378 real photographs has one object fully hidden behind a drawn mask, and the model is asked to name it: What is the primary object