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

Span-level Hallucination Detection research benchmarks · SOTA2 Research

Span-level Hallucination Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

LettuceDetect (test)

LettuceDetect-Qwen-2B

86.6F1 Score

12

Jul 2, 2026

HalluRAG (test)

80.56F1 Score

12

Feb 26, 2026

RagTruth-Avg (test)

76.63F1 Score

12

Feb 26, 2026

LettuceDetect Wikipedia (test)

LettuceDetect-Qwen-2B

81.7F1 Score

4

Jul 2, 2026

LettuceDetect Code-agent (test)

LettuceDetect-Qwen-2B

60.2F1 Score

4

Jul 2, 2026

RAGTruth (test)

LettuceDetect-Qwen-2B

57.4F1 Score

3

Jul 2, 2026

LettuceDetect-Qwen-2B

73.2F1 Score

3

Jul 2, 2026

LettuceDetect ALL (test)

LettuceDetect-Qwen-2B

68.9F1 Score

3

Jul 2, 2026

Popular datasets

LettuceDetect RAGTruth HalluRAG PsiloQA

Follow for updates

Join SOTA2 Contact