Salient Phrase Aware Dense Retrieval: Can a Dense Retriever Imitate a Sparse One?
Open paper
Benchmark
Model
Rank
Results
passage-retrieval-on-entityquestions
SPAR-PAQ
#2
Recall@20: 0.740
passage-retrieval-on-natural-questions
SPAR
#3
Precision@100: 88.8
Rank counts only results with a code link.