BEIR is a heterogeneous zero-shot benchmark showing BM25 as a robust baseline while re-ranking and late-interaction models perform best on average at higher cost, with dense and sparse models lagging in generalization.
Hersh, Kyle Lo, Kirk Roberts, Ian Soboroff, and Lucy Lu Wang
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2021 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models
BEIR is a heterogeneous zero-shot benchmark showing BM25 as a robust baseline while re-ranking and late-interaction models perform best on average at higher cost, with dense and sparse models lagging in generalization.