yqi19/BEAR-benchmark download history
yqi19/BEAR-benchmark is a visual question answering dataset on the Hugging Face Hub. In the last 30 days it was downloaded 1,020 times (398 in the last 7 days), and 3,453 times in total. It ranks #19,027 among datasets by monthly downloads.
BEAR Benchmark Data + runnable evaluation code for the BEAR benchmark. Each task folder holds its data (*.json + images/videos) and a run.sh. The shared runners and util/ live at the repository root. You run in two steps: Inference — a VLM answers every question → produces a final_*.