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

Hallucination robustness on TruthfulQA

48.76TruthfulQA Accuracy

Qwen3-1.7B

47.896848.120948.34548.5691May 28, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
48.76
2026.05
48.46
2026.05
48.4
2026.05
48.36
2026.05
48.35
2026.05
48.32
2026.05
48.3
2026.05
48.22
2026.05
48.16
2026.05
48.06
2026.05
48.06
2026.05
47.93