Passage Retrieval On Natural Questions

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Paper Title	Repository
DAR	Augmenting Document Representations for Dense Retrieval with Interpolation and Perturbation
DPR+RoBERTa-base-crossencoder-reranker	R2-D2: A Modular Baseline for Open-Domain Question Answering	-
ReAtt	Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer
DPR+ELECTRA-large-extreader-reranker	R2-D2: A Modular Baseline for Open-Domain Question Answering	-
BM25+RM3	Generation-Augmented Retrieval for Open-domain Question Answering
DPR	Dense Passage Retrieval for Open-Domain Question Answering
RocketQA	RocketQA: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answering
SPAR	Salient Phrase Aware Dense Retrieval: Can a Dense Retriever Imitate a Sparse One?
ANCE	Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
DPR-PAQ	Domain-matched Pre-training Tasks for Dense Retrieval

0 of 10 row(s) selected.