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/RAG

RAG

Reported on 87 benchmarks across 6 tasks · 2 papers · 6 SOTA

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

Natural Language Processing88 results

  • Fact VerificationonKILT: FEVER
    Accuracy· 2020-09-04
    86.31
    best: 89.55 (Re2G)
    SOTA
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Fact VerificationonKILT: FEVER
    KILT-AC· 2020-09-04
    53.45
    best: 78.53 (Re2G)
    SOTA
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Fact VerificationonKILT: FEVER
    R-Prec· 2020-09-04
    61.94
    best: 88.92 (Re2G)
    SOTA
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Fact VerificationonKILT: FEVER
    Recall@5· 2020-09-04
    75.55
    best: 92.52 (Re2G)
    SOTA
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Question AnsweringonNatural Questions
    EM· 2020-05-22
    44.5
    best: 64 (Atlas (full, Wiki-dec-2018 index))
    SOTA
    Retrieval-Augmented Generation for Knowledge-Intensive NLP TasksarXiv:2005.11401
  • Question AnsweringonWebQuestions
    EM· 2020-05-22
    45.2
    best: 84.6 (PoG-GPT4 (Tan et al., 2024))
    SOTA
    Retrieval-Augmented Generation for Knowledge-Intensive NLP TasksarXiv:2005.11401
  • Question AnsweringonKILT: ELI5
    F1· 2020-09-04
    14.51
    best: 27.13 (somebody)
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Question AnsweringonKILT: ELI5
    Rouge-L· 2020-09-04
    14.05
    best: 27.13 (RBG)
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Question AnsweringonKILT: ELI5
    F1· 2020-09-04
    14.51
    best: 27.13 (somebody)
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Question AnsweringonKILT: ELI5
    ROUGE-L· 2020-09-04
    14.05
    best: 24.53 (somebody)
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Open-Domain Question AnsweringonKILT: ELI5
    F1· 2020-09-04
    14.51
    best: 27.13 (somebody)
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Open-Domain Question AnsweringonKILT: ELI5
    ROUGE-L· 2020-09-04
    14.05
    best: 24.53 (somebody)
    KILT: a Benchmark for Knowledge Intensive Language TasksarXiv:2009.02252
  • Question AnsweringonTriviaQA
    EM· 2020-05-22
    56.1
    best: 87.5 (Claude 2 (few-shot, k=5))
    Retrieval-Augmented Generation for Knowledge-Intensive NLP TasksarXiv:2005.11401
  • Fact VerificationonFEVER
    Accuracy· uses extra data· 2020-05-22
    72.5
    best: 79.47 (ProoFVer-SB)
    Retrieval-Augmented Generation for Knowledge-Intensive NLP TasksarXiv:2005.11401
  • Question AnsweringonKILT: TriviaQA
    EM
    71.27
    best: 76.27 (Re2G)
  • Question AnsweringonKILT: TriviaQA
    F1
    75.88
    best: 81.4 (Re2G)
  • Question AnsweringonKILT: TriviaQA
    KILT-EM
    38.13
    best: 57.91 (Re2G)
  • Question AnsweringonKILT: TriviaQA
    KILT-F1
    40.15
    best: 61.78 (Re2G)
  • Question AnsweringonKILT: TriviaQA
    R-Prec
    48.68
    best: 72.68 (Re2G)
  • Question AnsweringonKILT: TriviaQA
    Recall@5
    57.13
    best: 76.36 (intersect)
  • Question AnsweringonKILT: Natural Questions
    EM
    44.39
    best: 53.74 (intersect)
  • Question AnsweringonKILT: Natural Questions
    F1
    52.35
    best: 62.24 (intersect)
  • Question AnsweringonKILT: Natural Questions
    KILT-EM
    32.69
    best: 43.56 (Re2G)
  • Question AnsweringonKILT: Natural Questions
    KILT-F1
    37.91
    best: 49.8 (Re2G)
  • Question AnsweringonKILT: Natural Questions
    R-Prec
    59.49
    best: 70.78 (Re2G)
  • Question AnsweringonKILT: Natural Questions
    Recall@5
    67.06
    best: 76.63 (Re2G)
  • Question AnsweringonKILT: HotpotQA
    EM
    26.97
    best: 40.46 (intersect)
  • Question AnsweringonKILT: HotpotQA
    F1
    36.03
    best: 51.44 (intersect)
  • Question AnsweringonKILT: HotpotQA
    KILT-EM
    3.21
    best: 18.06 (intersect)
  • Question AnsweringonKILT: HotpotQA
    KILT-F1
    4.1
    best: 21.42 (intersect)
  • Question AnsweringonKILT: HotpotQA
    R-Prec
    30.59
    best: 58.83 (intersect)
  • Question AnsweringonKILT: HotpotQA
    Recall@5
    12.59
    best: 51.03 (intersect)
  • Question AnsweringonKILT: ELI5
    KILT-F1
    1.79
    best: 3 (somebody)
  • Question AnsweringonKILT: ELI5
    KILT-RL
    1.69
    best: 2.62 (somebody)
  • Question AnsweringonKILT: ELI5
    R-Prec
    11
    best: 18.33 (TABi)
  • Question AnsweringonKILT: ELI5
    Recall@5
    22.92
    best: 28.21 (TABi)
  • Entity LinkingonKILT: WNED-WIKI
    Accuracy
    48.07
    best: 87.44 (GENRE)
  • Entity LinkingonKILT: WNED-WIKI
    KILT-AC
    48.07
    best: 87.44 (GENRE)
  • Entity LinkingonKILT: WNED-WIKI
    R-Prec
    48.07
    best: 88.12 (chriskuei)
  • Entity LinkingonKILT: WNED-WIKI
    Recall@5
    48.07
    best: 95.62 (chriskuei)
  • Entity LinkingonKILT: AIDA-YAGO2
    Accuracy
    72.62
    best: 89.85 (GENRE)
  • Entity LinkingonKILT: AIDA-YAGO2
    KILT-AC
    72.62
    best: 89.85 (GENRE)
  • Entity LinkingonKILT: AIDA-YAGO2
    R-Prec
    72.62
    best: 89.98 (chriskuei)
  • Entity LinkingonKILT: AIDA-YAGO2
    Recall@5
    72.62
    best: 94.85 (chriskuei)
  • Entity LinkingonKILT: WNED-CWEB
    Accuracy
    47.61
    best: 71.22 (GENRE)
  • Entity LinkingonKILT: WNED-CWEB
    KILT-AC
    47.61
    best: 71.22 (GENRE)
  • Entity LinkingonKILT: WNED-CWEB
    R-Prec
    47.61
    best: 71.22 (GENRE)
  • Entity LinkingonKILT: WNED-CWEB
    Recall@5
    47.61
    best: 81.76 (BLINK)
  • Slot FillingonKILT: T-REx
    Accuracy
    59.2
    best: 87.68 (Re2G)
  • Slot FillingonKILT: T-REx
    F1
    62.96
    best: 89.93 (Re2G)
  • Slot FillingonKILT: T-REx
    KILT-AC
    23.12
    best: 75.84 (Re2G)
  • Slot FillingonKILT: T-REx
    KILT-F1
    23.94
    best: 77.05 (Re2G)
  • Slot FillingonKILT: T-REx
    R-Prec
    28.68
    best: 81.9 (TABi)
  • Slot FillingonKILT: T-REx
    Recall@5
    33.04
    best: 89.36 (TABi)
  • Slot FillingonKILT: Zero Shot RE
    Accuracy
    44.74
    best: 74.63 (single ngram)
  • Slot FillingonKILT: Zero Shot RE
    F1
    49.95
    best: 79.66 (single ngram)
  • Slot FillingonKILT: Zero Shot RE
    KILT-AC
    36.83
    best: 73.2 (single ngram)
  • Slot FillingonKILT: Zero Shot RE
    KILT-F1
    39.91
    best: 78.12 (single ngram)
  • Slot FillingonKILT: Zero Shot RE
    R-Prec
    53.73
    best: 98.49 (KGI_1)
  • Slot FillingonKILT: Zero Shot RE
    Recall@5
    59.52
    best: 99.34 (single ngram)
  • Open-Domain Question AnsweringonKILT: TriviaQA
    EM
    71.27
    best: 76.27 (Re2G)
  • Open-Domain Question AnsweringonKILT: TriviaQA
    F1
    75.88
    best: 81.4 (Re2G)
  • Open-Domain Question AnsweringonKILT: TriviaQA
    KILT-EM
    38.13
    best: 57.91 (Re2G)
  • Open-Domain Question AnsweringonKILT: TriviaQA
    KILT-F1
    40.15
    best: 61.78 (Re2G)
  • Open-Domain Question AnsweringonKILT: TriviaQA
    R-Prec
    48.68
    best: 72.68 (Re2G)
  • Open-Domain Question AnsweringonKILT: TriviaQA
    Recall@5
    57.13
    best: 76.36 (intersect)
  • Open-Domain Question AnsweringonKILT: Natural Questions
    EM
    44.39
    best: 53.74 (intersect)
  • Open-Domain Question AnsweringonKILT: Natural Questions
    F1
    52.35
    best: 62.24 (intersect)
  • Open-Domain Question AnsweringonKILT: Natural Questions
    KILT-EM
    32.69
    best: 43.56 (Re2G)
  • Open-Domain Question AnsweringonKILT: Natural Questions
    KILT-F1
    37.91
    best: 49.8 (Re2G)
  • Open-Domain Question AnsweringonKILT: Natural Questions
    R-Prec
    59.49
    best: 70.78 (Re2G)
  • Open-Domain Question AnsweringonKILT: Natural Questions
    Recall@5
    67.06
    best: 76.63 (Re2G)
  • Open-Domain Question AnsweringonKILT: HotpotQA
    EM
    26.97
    best: 40.46 (intersect)
  • Open-Domain Question AnsweringonKILT: HotpotQA
    F1
    36.03
    best: 51.44 (intersect)
  • Open-Domain Question AnsweringonKILT: HotpotQA
    KILT-EM
    3.21
    best: 18.06 (intersect)
  • Open-Domain Question AnsweringonKILT: HotpotQA
    KILT-F1
    4.1
    best: 21.42 (intersect)
  • Open-Domain Question AnsweringonKILT: HotpotQA
    R-Prec
    30.59
    best: 58.83 (intersect)
  • Open-Domain Question AnsweringonKILT: HotpotQA
    Recall@5
    12.59
    best: 51.03 (intersect)
  • Open-Domain Question AnsweringonKILT: ELI5
    KILT-F1
    1.79
    best: 3 (somebody)
  • Open-Domain Question AnsweringonKILT: ELI5
    KILT-RL
    1.69
    best: 2.62 (somebody)
  • Open-Domain Question AnsweringonKILT: ELI5
    R-Prec
    11
    best: 18.33 (TABi)
  • Open-Domain Question AnsweringonKILT: ELI5
    Recall@5
    22.92
    best: 28.21 (TABi)
  • Open-Domain DialogonKILT: Wizard of Wikipedia
    F1
    13.11
    best: 19.19 (Hindsight)
  • Open-Domain DialogonKILT: Wizard of Wikipedia
    KILT-F1
    8.75
    best: 13.39 (Hindsight)
  • Open-Domain DialogonKILT: Wizard of Wikipedia
    KILT-RL
    7.59
    best: 11.92 (Hindsight)
  • Open-Domain DialogonKILT: Wizard of Wikipedia
    R-Prec
    57.75
    best: 64.79 (chriskuei)
  • Open-Domain DialogonKILT: Wizard of Wikipedia
    ROUGE-L
    11.57
    best: 17.06 (Hindsight)
  • Open-Domain DialogonKILT: Wizard of Wikipedia
    Recall@5
    74.61
    best: 82.15 (chriskuei)