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  new activity 4 days ago

BeIR/fiqa:VORTEXRAG: 7-Layer RAG — EM 74.8 on QA benchmarks, solves Semantic Drift [open source]

nthakur  new activity 4 days ago

BeIR/hotpotqa:VORTEXRAG: 7-Layer RAG — EM 74.8 on QA benchmarks, solves Semantic Drift [open source]

nthakur  new activity 4 days ago

BeIR/nq:VORTEXRAG: 7-Layer RAG — EM 74.8 on QA benchmarks, solves Semantic Drift [open source]

View all activity

Team members 2