Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Truthful Question Answering on TruthfulQA o=5 (f_clean)
Loading...
55.5
Accuracy (Exact)
baseline
27.628
34.864
42.1
49.336
Jan 27, 2026
Accuracy (Exact)
Accuracy (Semantic-level)
Accuracy (Domain-level)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (Exact)
Accuracy (Semantic-level)
Accuracy (Domain-level)
baseline
Model=Qwen 2.5, Access=–
2026.01
55.5
53.1
57.2
baseline
Model=LLaMA-3, Access=–
2026.01
47.4
46.7
45.2
baseline
Model=Mistral, Access=–
2026.01
28.7
24.9
26.9
Feedback
Search any
task
Search any
task
