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

Hallucination on TruthfulQA

75.76Score

ICL

32.672843.858955.04566.2311Dec 8, 2025Dec 18, 2025Dec 28, 2025Jan 7, 2026Jan 17, 2026Jan 27, 2026Feb 7, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
75.76
2026.02
72.22
2026.02
71.71
2026.02
71.06
2026.02
70.84
2026.02
70.63
2026.02
70.54
2026.02
70.44
2026.02
67.33
2026.02
65.5
2026.02
64.58
2026.02
64.13
2026.02
62.16
2026.02
59.62
2026.02
56.45
2026.02
54.94
2026.02
54.02
2025.12
52.85
2025.12
52.12
2025.12
51.77
2025.12
51.73
2026.02
50.8
2025.12
48.93
2025.12
48.74
2025.12
48.51
2025.12
48.38
2025.12
48.32
2025.12
47.64
2025.12
46.93
2026.02
46.18
2025.12
46.05
2025.12
46
2025.12
45.95
2025.12
45.92
2025.12
43.93
2025.12
43.67
2025.12
43.08
2025.12
42.59
2025.12
42.56
2025.12
41.46
2026.02
35.38
2026.02
34.33