datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
HVMap-dataset
HVMap Dataset
HVMap Dataset is a BEIR/MTEB-style text-retrieval benchmark package for
federated vector retrieval across heterogeneous vector databases. It contains
the retrieval corpus, mixed-domain queries, and qrels used by the HVMap
cross-domain benchmark described in Federated Vector Retrieval across
Heterogeneous VectorDBs.
Repository Layout
Path
Description
data/corpus/*.jsonl
Search corpus shards derived from FIQA, BioASQ, SciDocs, and SciFact.… See the full description on the dataset page: https://huggingface.co/datasets/snu-aidas/HVMap-dataset.affine-combined-hvm-dpo