1 paper
Yunye Gong, Robik Shrestha, Jared Claypoole +4
We propose a novel VQA dataset, BloomVQA, to facilitate comprehensive evaluation of large vision-language models on comprehension tasks. Unlike current benchmarks that often focus…