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.
For example, for scientific publication we have the title, the abstract, the document body, the authors list, and the journal name
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.