Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Hallucination Detection on RAGTruth Span-level, leakage-clean protocol

0.702AUC

GASP+base (trained)

0.515840.564170.61250.66083Jul 5, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.07
0.702
2026.07
0.681
2026.07
0.678
2026.07
0.676
2026.07
0.673
2026.07
0.672
2026.07
0.657
2026.07
0.645
2026.07
0.635
2026.07
0.615
2026.07
0.565
2026.07
0.548
2026.07
0.547
2026.07
0.536
2026.07
0.523