BEIR

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.

Recent Activity

nthakur  authored a paper about 2 months ago

MIRAGE-Bench: Automatic Multilingual Benchmark Arena for Retrieval-Augmented Generation Systems

nthakur  authored a paper 4 months ago

Ragnarök: A Reusable RAG Framework and Baselines for TREC 2024 Retrieval-Augmented Generation Track

nthakur  authored a paper 12 months ago

Resources for Brewing BEIR: Reproducible Reference Models and an Official Leaderboard

View all activity

Team members 2