NagaYu/sieve-eval download history

NagaYu/sieve-eval is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 109 times (16 in the last 7 days), and 217 times in total. It ranks #141,440 among datasets by monthly downloads.

Sieve Eval A small English/Japanese benchmark for auditing RAG context: is a retrieved chunk unsafe to pass to an LLM, and when a RAG exchange goes wrong, did retrieval or generation fail? It was built to measure Sieve, a deterministic, rule-based RAG context auditor (engine · in-brow

Models trained on sieve-eval

1 models list it as training data.

Spaces using sieve-eval

Open NagaYu/sieve-eval on Hugging Face

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