Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Open-domain Question Answering research benchmarks – Page 4 · SOTA2 Research

Open-domain Question Answering

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

SQuAD v1.1 (test)

75.1Acc@20

9

Feb 26, 2026

NaturalQuestions (test)

Reader-to-Retriever Distillation

54.4Top-1 EM

9

Feb 26, 2026

TriviaQA hidden (test)

Fusion-in-Decoder

80.1EM

9

Feb 26, 2026

Natural Questions Open (dev)

36.8EM

9

Feb 26, 2026

17.5Exact Match (EM)

8

Jun 16, 2026

59.36Exact Match (EM)

8

Apr 28, 2026

Self-RAG + QREAM-ICL

71.4Accuracy

8

Apr 21, 2026

Top 5 documents

517.6Words

8

Feb 27, 2026

Hotpot QA (val)

Llama2-7b CBQA w. Mixed Training

24.6EM

8

Feb 26, 2026

Llama2-7b CBQA w. Mixed Training + Recitation

30.6EM

8

Feb 26, 2026

TriviaQA (TQA) (dev)

72.7EM

8

Feb 26, 2026

Top 5 documents

514.6Word Score

8

Feb 27, 2026

52.51Trivia QA Score

7

Jun 2, 2026

TriviaQA, NQ Open, WebQS Foundation-Preservation Suite

52.51TriviaQA Score

7

Jun 2, 2026

Original (Unsafe)

72.82EM Accuracy

7

Mar 19, 2026

Original (Unsafe)

59.62EM Accuracy

7

Mar 19, 2026

EntityQuestions (ENTITYQ) (test)

DensePhrases_NQ + TOUR_hard

28.3EM

7

Feb 26, 2026

Natural Questions (NQ) few-shot (test)

Codex + REPLUG LSR

45.5EM

7

Feb 26, 2026

40.7Accuracy

7

Feb 26, 2026

TriviaQA (TQA) Wiki (test)

75Exact Match

7

Feb 26, 2026

OpenSQuAD 1.1 (test)

Multi-Passage BERT

53EM

7

Feb 26, 2026

ConSA (head-wise, single-layer)

57.15Accuracy

6

Jun 17, 2026

Seven-benchmark

46.9NQ

6

Jun 2, 2026

Natural Questions abundant

37EM & F1 Score

6

Feb 26, 2026

Natural Questions (medium)

36.5F1 Score

6

Feb 26, 2026

Page 4 of 7

Popular datasets

TriviaQA Natural Questions HotpotQA MuSiQue NQ

Follow for updates

Join SOTA2 Contact