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

Hallucination Detection research benchmarks – Page 17 · SOTA2 Research

Hallucination Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

100AUC

3

Feb 26, 2026

84AUC

3

Feb 26, 2026

Legal Contract QA

0.94AUC

3

Feb 26, 2026

BEGIN Out-of-Distribution

GPT-4 (Internal)

54.3Macro F1

3

Feb 26, 2026

83.2Adversarial Accuracy (POPE)

2

Jun 11, 2026

68.7AUROC

2

May 28, 2026

6292 OR-expert-labeled artifacts natural benchmark

61.7Hallucination F1

2

May 13, 2026

0.6AUROC (Token Entropy)

2

Apr 21, 2026

VideoHallucer Object-Relation, Temporal, Semantic Detail, Factual, Non-factual

LLaVA-NeXT-Video-7B + DTR

52.5Object-Relation Accuracy

2

Apr 15, 2026

57.3F1 Score

2

Mar 23, 2026

72.9Accuracy

2

Mar 13, 2026

RSHalluEval 1.0 (test)

1.24ES_IA

2

Feb 26, 2026

0.883AUC ROC

2

Feb 26, 2026

0.898AUC ROC

2

Feb 26, 2026

DiaHalu (sampled)

87.7Precision (Class 0)

1

May 13, 2026

Pretrained factual knowledge 196 facts

Geometric margin

1Margin AUROC

1

May 8, 2026

Parametric Memory (PM) LoRA-installed 1600 entities

Geometric margin

99.3Margin AUROC

1

May 8, 2026

HaluEval Dialogue (test)

Directional Grounding Index (Γ)

0.287Groundedness (Gamma)

1

Feb 26, 2026

Financial QA dataset

0.89AUC

1

Feb 26, 2026

HaluEval ChatGPT (test)

94.5Coverage

1

Feb 26, 2026

Financial QA n=200 balanced samples (test)

—

—Primary metric

0

Feb 26, 2026

Various benchmarks

—

—Primary metric

0

Feb 26, 2026

—

—Primary metric

0

Feb 26, 2026

Page 17 of 17

Popular datasets

RAGTruth HaluEval TriviaQA SQuAD CoQA

Follow for updates

Join SOTA2 Contact