qbao775/naf-bench download history

qbao775/naf-bench is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 256 times (66 in the last 7 days), and 445 times in total. It ranks #55,533 among datasets by monthly downloads.

NAF-Bench Does an LLM follow a specified reading of default negation (SLDNF / well-founded / credulous / skeptical stable-model)? Solver-certified and generated fresh so it stays contamination-free. Primary metric: JOINT accuracy — a program counts only if all four specified readings

Open qbao775/naf-bench on Hugging Face

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