soumikmahato/diverse-websearch-3.5k

Diverse WebSearch 3.5k Diverse WebSearch 3.5K is a small web research dataset containing 3,500 diverse web pages collected from search results. Each row contains a source URL, extracted markdown content, a concise summary, and image URLs found on the page. Dataset Details This dataset is intended for learning and experimentation with: webpage summarization retrieval-augmented generation search result understanding document cleaning synthetic QA generation dataset… See the full description on the dataset page: https://huggingface.co/datasets/soumikmahato/diverse-websearch-3.5k.