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

LLM Judgment on HotpotQA

67.24HSPP

Self-Consistency

-1.077616.658734.39552.1313Oct 9, 2025
Updated 3mo ago

Evaluation Results

MethodLinks
2025.10
67.24
2025.10
63.58
2025.10
62.93
2025.10
62.93
2025.10
60.96
2025.10
53.23
2025.10
50.43
2025.10
45.47
2025.10
44.51
2025.10
40.61
2025.10
40.52
2025.10
39.7
2025.10
39.16
2025.10
39.01
2025.10
37.89
2025.10
34.7
2025.10
34.39
2025.10
21.12
2025.10
3.28
2025.10
3.09
2025.10
3.09
2025.10
2.84
2025.10
1.55