TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/DPR

DPR

Reported on 7 benchmarks across 2 tasks · 2 papers · 4 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing7 results

  • Question AnsweringonNatural Questions
    EM· 2020-04-10
    41.5
    best: 64 (Atlas (full, Wiki-dec-2018 index))
    SOTA
    Dense Passage Retrieval for Open-Domain Question AnsweringarXiv:2004.04906
  • Question AnsweringonNaturalQA
    EM· 2020-04-10
    41.5
    SOTA
    Dense Passage Retrieval for Open-Domain Question AnsweringarXiv:2004.04906
  • Information RetrievalonNatural Questions
    Precision@100· 2020-04-10
    86
    best: 90.4 (ReAtt)
    SOTA
    Dense Passage Retrieval for Open-Domain Question AnsweringarXiv:2004.04906
  • Information RetrievalonNatural Questions
    Precision@20· 2020-04-10
    79.4
    best: 86 (ReAtt)
    SOTA
    Dense Passage Retrieval for Open-Domain Question AnsweringarXiv:2004.04906
  • Information RetrievalonMSMARCO (BEIR)
    nDCG@10· 2021-04-17
    0.177
    best: 0.413 (BM25+CE)
    BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval ModelsarXiv:2104.08663
  • Question AnsweringonWebQuestions
    EM· 2020-04-10
    42.4
    best: 84.6 (PoG-GPT4 (Tan et al., 2024))
    Dense Passage Retrieval for Open-Domain Question AnsweringarXiv:2004.04906
  • Question AnsweringonTriviaQA
    EM· 2020-04-10
    56.8
    best: 87.5 (Claude 2 (few-shot, k=5))
    Dense Passage Retrieval for Open-Domain Question AnsweringarXiv:2004.04906