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

LongFact dataset and benchmarks · SOTA2 Research

Research Datasets

LongFact

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Uncertainty Quantification

LongFact

-0.017PCC

32

Factuality Hallucination Evaluation

LongFact (test)

100Response Score

30

Factuality Hallucination

LongFact

23.5Facts Score

30

Long-form Question Answering

LongFact FullP (100% polluted)

88.93VeriScore F1@k

26

Long-form Question Answering

LongFact MixP (50% polluted)

90.33VeriScore F1@k

26

Long-form Factuality

LongFact

78.4R@64

18

Long-form Factual Generation

LongFact

84.2Fact Recall (FR) - Science

14

Long-form Retrieval-Augmented Generation

LongFact

247.3Information Density (Sci.)

14

Factual Text Generation

LongFact Objects

0.426AURC

14

Long-form generation factuality and uncertainty estimation

LongFact (test)

91.5Factuality Score

14

Long-form Question Answering

LongFact

75.9VeriScore F1

14

Factuality and Hallucination Detection

LongFact Concepts

99.7Overall Score

9

Factuality and Hallucination Detection

LongFact Objects

99.2Factuality Score

9

Long-form factuality evaluation

LongFact

90.2Accuracy

7

Claim-level specificity control

LongFact full

11,705Claims Emitted

6

Factuality Evaluation

LongFact

38.6Precision

6

Long-form Generation

LongFact

100Response Rate

4

Hallucination Detection

LongFact-Aug (test)

0.94AUC

4

Claim-level specificity control

LongFact pilot

724Claims Emitted

3

Popular tasks

Long-form Question Answering Claim-level specificity control Factuality and Hallucination Detection Long-form generation Hallucination Detection

Follow for updates

Join SOTA2 Contact