Faithfulness Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
2.3AUC π-Soft-NS
67
Apr 16, 2026
2.25AUC π-Soft-NS
67
Apr 16, 2026
74.9F1 Score
42
Apr 14, 2026
1.293Insertion Score
40
May 19, 2026
87Comp Score
40
Feb 27, 2026
61.41Deletion Score
30
Mar 17, 2026
37AUC π-Soft-NS
27
Feb 27, 2026
0.563AUC π-Soft (NS)
27
Feb 27, 2026
57.2AUC π-Soft-NS
27
Feb 27, 2026
92.2Rank Accuracy (RA)
25
Mar 23, 2026
2Rate of Label Changes
24
Apr 1, 2026
4.5Rate of Label Changes
24
Apr 1, 2026
5.5Rate of Label Changes
24
Apr 1, 2026
81.03ADCC
24
Mar 17, 2026
73CODE
20
Mar 17, 2026
98.94NAR Score
18
Mar 13, 2026
20Rate of Label Changes
12
Apr 1, 2026
18Rate of Label Changes
12
Apr 1, 2026
28Rate of Label Changes
12
Apr 1, 2026
0.282SummaC Score
12
Feb 26, 2026
40.71SummaC
12
Feb 26, 2026
38.5SummaC
12
Feb 26, 2026
53.58SummaC
12
Feb 26, 2026
39.06SummaC
12
Feb 26, 2026
34.39SummaC
12
Feb 26, 2026