visual-question-answering-benchmark
A Python framework for collecting multilingual, culturally aligned question-answering datasets from web search to support LLM evaluation and benchmarking.
