cybersec-retrieval-benchmark

简介

CyberSec Retrieval Benchmark Multi-task retrieval benchmark for cybersecurity embedding and reranking models. Configs Config Contents queries Evaluation queries with ask labels corpus Document passages per task qrels Relevance judgments ( elevance=1 for gold docs) python from datasets import load_dataset queries = load_dataset("alirezaaminzadeh/cybersec-retrieval-benchmark", "queries", split="train") corpus =… See the full description on the dataset page: https://huggingface.co/datasets/alirezaaminzadeh/cybersec-retrieval-benchmark.