Faithfulness Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
27.06AOPC Completeness Gap
4
Jun 30, 2026
3.31AOPC Completeness Gap
4
Jun 30, 2026
13.21AOPC Completeness Gap
4
Jun 30, 2026
0.535Kendall's tau (τ)
3
Apr 13, 2026
100Sufficiency
3
Mar 12, 2026
0.581Correlation
2
Feb 26, 2026
0.507Correlation
2
Feb 26, 2026
0.795Correlation (Human Judgment)
2
Feb 26, 2026
0.82Faithfulness Correlation
2
Feb 26, 2026
—
—AUC π-Soft-NS
0
Feb 26, 2026
—
—AUC π-Soft-NS
0
Feb 26, 2026
—
—AUC π-Soft-NS
0
Feb 26, 2026
—
—AUC π-Soft-NS
0
Feb 26, 2026