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

Multi-hop Reasoning on HotpotQA

67.19AUROC

CoT-UQ

47.253252.429157.60562.7809Feb 24, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.02
67.19
2025.02
66.56
2025.02
66.29
2025.02
64.37
2025.02
64.03
2025.02
63.1
2025.02
62.39
2025.02
62.01
2025.02
61.07
2025.02
60.63
2025.02
58.59
2025.02
58.34
2025.02
58.24
2025.02
57.33
2025.02
57.2
2025.02
57.1
2025.02
56.51
2025.02
56.27
2025.02
55.62
2025.02
54.63
2025.02
54.33
2025.02
53.73
2025.02
53.57
2025.02
51.13
2025.02
50.18
2025.02
48.02