Hallucination Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
79.1Accuracy
23
Feb 27, 2026
19.4Average CI Score
21
Jul 7, 2026
91.1Accuracy
21
May 15, 2026
55.2CS Score
21
Mar 18, 2026
69.5HR^D
21
Feb 26, 2026
12.1HALL%
21
Feb 26, 2026
1.019CHAIRs
20
May 19, 2026
50.9CS Score
20
May 12, 2026
89.4Score
20
Apr 3, 2026
80.8Score S
19
Mar 19, 2026
54.2Score
19
Feb 26, 2026
61.6TBH Score
18
Jun 4, 2026
89.3Accuracy
16
May 14, 2026
59.5CHAIRs
16
Apr 10, 2026
195Existence Score
16
Mar 27, 2026
65.9Multi-Object Paired Accuracy
16
Mar 19, 2026
80Multi-obj Acc (Paired)
16
Mar 19, 2026
671.6Accuracy
15
Apr 29, 2026
489.89MVH Score
15
Mar 26, 2026
71Binary Accuracy
15
Mar 26, 2026
73.66Binary Acc
15
Mar 26, 2026
91.8Accuracy (Random)
15
Mar 18, 2026
93.76Accuracy (Random)
15
Mar 18, 2026
91.6Accuracy (Rand.)
15
Mar 18, 2026
200Existence
15
Mar 18, 2026