You can deploy this model and start earning money today!

Dataset Card for BEIR Benchmark

scidocs is one of the datasets from the Citation Prediction task within BEIR, measuring cited scientific article retrieval for a given scientific title.

Dataset Summary

BEIR is a heterogeneous benchmark built from 18 diverse datasets representing 9 information retrieval tasks.

Fact-checking: FEVER, Climate-FEVER, SciFact Question-Answering: NQ, HotpotQA, FiQA-2018 Bio-Medical IR: TREC-COVID, BioASQ, NFCorpus News Retrieval: TREC-NEWS… See the full description on the dataset page: https://huggingface.co/datasets/BeIR/scidocs.