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

Hallucination Assessment on HallusionBench

71.6Answer Accuracy (aAcc)

ResDec

33.983243.749153.51563.2809Jan 24, 2026Jan 25, 2026Jan 27, 2026Jan 28, 2026Jan 30, 2026Jan 31, 2026Feb 2, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.02
71.64947.1--
2026.01
69.19----
2026.02
69.145.843.4--
2026.02
68.743.441.5--
2026.02
68.741.339.8--
2026.01
68.45----
68.35----
2026.01
68.24----
67.93----
2026.01
67.82----
2026.01
67.61----
2026.01
67.61----
2026.02
67.639.838.7--
67.51----
2026.02
66.939.540.8--
2026.02
65.133.4147.1156.98-
2026.02
63.5130.1143.3554.88-
2026.02
60.0527.9136.7153.49-
2026.02
58.924.1840.1754.42-
2026.02
58.4623.0731.549.3-
2026.02
55.4521.7628.6145.81-
2026.02
48.8919.3427.7640-
2026.02
46.510.211.3--
2026.02
45.99.29.7--
2026.02
45.39.510.1--
2026.02
45.112.718.2--
2026.02
43.88.68.4--
2026.02
43.38.58.1--
2026.02
42.98.317.2--
2026.02
42.79.317.5--
2026.02
42.5917.9--
2026.02
42.38.29.7--
2026.02
41.58.117.9--
2026.02
41.59.315.7--
2026.02
41.25.516.2--
2026.02
41.111.413.9--
2026.02
417.57.2--
2026.02
38.28.413.9--
2026.02
35.438.5712.4331.86-
2026.03
----68.9
2026.03
----66.8
2026.03
----69.7
2026.03
----67
2026.03
----69.8
2026.03
----62.8
2026.03
----68
2026.03
----66.1
2026.03
----69.5
2026.03
----65.7
2026.03
----70
2026.03
----65.2
2026.03
----70
2026.03
----69.1