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

HotpotQA dataset and benchmarks – Page 20 · SOTA2 Research

Research Datasets

HotpotQA

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Question Answering

HotpotQA

63.1AUROC

6

Open-domain Question Answering

HotpotQA (abundant)

45.7F1 Score

6

Open-domain Question Answering

HotpotQA (medium)

43.9F1 Score

6

Open-domain Question Answering

HotpotQA (scarce)

41.9EM & F1 Score

6

Multi-hop Question Answering

HotpotQA blind (test)

72.69Answer EM

6

Question Answering

HotpotQA subset of 500 questions (dev)

31.4EM (ReAct)

6

HotpotQA (train)

54.4Recall@2

6

Question Generation

HotpotQA (test)

0.211BLEU-3

6

Multi-hop Reading Comprehension

HotpotQA distractor (test)

69.76F1 Score

6

Question Answering

HotpotQA

32.14Relevance

5

Question Answering

HotpotQA

79.36F1 Score

5

Multi-Hop Reasoning

HotpotQA

0.722Accuracy

5

Document Reranking

HotpotQA

81.1NDCG@5

5

HotpotQA 37k-node

0Blind Harm Count

5

Question Answering

HotpotQA

0Build Cost (Tokens)

5

Offline construction cost evaluation

HotPotQA

542Output Tokens per 1k Items

5

Multi-hop Question Answering

HotpotQA Medium-scale corpus

62.3EM

5

Multi-hop Question Answering

HotpotQA

62.2Correctness

5

Knowledge corruption attack

HotpotQA

100ASR

5

HotpotQA

81.3R@5

5

Question Answering

HotpotQA

47.8EM

5

Question Answering

HotpotQA

61.3SubEM

5

Dense Retrieval

HotpotQA

0.812NDCG@10

5

Question Answer

HotPotQA (HPQA) distractor Open-book

87.9Recall

5

Dense passage retrieval

HotpotQA

95.85Hit@5

5

Page 20 of 27

Popular tasks

Question Answering Multi-hop Question Answering Retrieval Open-domain Question Answering Multi-hop Reasoning

Follow for updates

Join SOTA2 Contact