jayshah5696/tiny-evidence-analyst-benchmark download history

jayshah5696/tiny-evidence-analyst-benchmark is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 86 times (13 in the last 7 days), and 86 times in total. It ranks #127,866 among datasets by monthly downloads.

Tiny Evidence Analyst benchmark This dataset tests short answers with passage-level citations. It has a synthetic evaluator-attack layer plus web_5page and rag_10chunk real-data families. The synthetic layer contains 12 cases and 24 hand-labeled candidate outputs. web_5page contains 1

Open jayshah5696/tiny-evidence-analyst-benchmark on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.