Hallucination assessment
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
10.6CHAIR_s
56
Jun 1, 2026
5.6CHAIR
44
Jun 11, 2026
71.6Answer Accuracy (aAcc)
39
Apr 23, 2026
2.6Mention Hallucination Rate
39
Feb 26, 2026
3.85Bingo Score
21
Apr 14, 2026
4.27MMHalu Score
21
Apr 14, 2026
0.979HF Information Accuracy
21
Feb 26, 2026
72.6Response Rate
20
Feb 26, 2026
52.8CS
12
May 26, 2026
2.71GPT-Hall Score
11
Feb 26, 2026
12.2CHAIR_I
6
Feb 26, 2026
10.7CHAIR Instance Score
3
Feb 26, 2026