BEIR

university

AI & ML interests

BEIR (Benchmarking IR) consists of a homogenous benchmark for diverse sentence or passage level IR tasks. It provides a common and easy framework for the cross-domain evaluation of your retrieval models.

models 3

datasets 49

BeIR/bioasq-generated-queries

Viewer • Updated Oct 23, 2022 • 3.01M • 87 • 2

BeIR/cqadupstack-qrels

Viewer • Updated Oct 23, 2022 • 23.7k • 75

BeIR/cqadupstack-generated-queries

Viewer • Updated Oct 23, 2022 • 1.35M • 106 • 1

BeIR/hotpotqa-generated-queries

Viewer • Updated Oct 23, 2022 • 5.23M • 51 • 1

BeIR/nq-generated-queries

Viewer • Updated Oct 23, 2022 • 7.87M • 73 • 1

BeIR/quora-generated-queries

Viewer • Updated Oct 23, 2022 • 523k • 46

BeIR/signal1m-generated-queries

Viewer • Updated Oct 23, 2022 • 8.42M • 49

BeIR/robust04-generated-queries

Viewer • Updated Oct 23, 2022 • 1.56M • 82 • 1

BeIR/webis-touche2020-generated-queries

Viewer • Updated Oct 23, 2022 • 1.12M • 73

BeIR/trec-news-generated-queries

Viewer • Updated Oct 23, 2022 • 1.76M • 106 • 3

AI & ML interests

Team members 2

models 3 Sort: Recently updated

datasets 49 Sort: Recently updated

models 3

datasets 49