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

Question Answering on TruthfulQA (MC1, MC2, Bleu, Bleurt)

36.5MC1 Score

ITI

18.123222.894127.66532.4359Aug 14, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.08
36.554.2943.2266.3
2025.08
34.9156.3252.3266.75
2025.08
34.8253.2352.1863.26
2025.08
34.7253.2753.174.9
2025.08
31.3252.6250.6171.39
2025.08
30.0749.6949.3966.01
2025.08
29.8348.6952.5771.15
2025.08
29.5849.5152.0864.06
2025.08
29.4551.1249.9569.2
2025.08
29.2949.6743.8155.31
2025.08
29.1548.9252.8572.25
2025.08
28.7752.0553.8562.27
2025.08
28.4447.2548.3558.97
2025.08
28.4147.5545.4261.54
2025.08
28.3746.2148.3560.44
2025.08
27.4745.6346.8957.88
2025.08
26.8448.3349.7966.63
2025.08
26.3845.4149.6365.28
2025.08
25.4548.3849.4662.62
2025.08
23.0849.5151.7272.53
2025.08
21.9249.2344.1157.64
2025.08
18.8336.5441.5654.52