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

Multi-hop reasoning on MuSiQue Open Book (test)

75.05Accuracy / F1

DeCoRe (static)

36.840446.760256.6866.5998Jun 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
75.05
2026.06
74.64
2026.06
74.43
2026.06
74.27
2026.06
74.27
2026.06
74.27
2026.06
74.02
2026.06
73.02
2026.06
72.49
2026.06
71.7
2026.06
69.92
2026.06
69.92
2026.06
69.84
2026.06
69.76
2026.06
69.01
2026.06
68.68
2026.06
68.64
2026.06
68.56
2026.06
68.14
2026.06
68.14
2026.06
67.89
2026.06
66.61
2026.06
61.23
2026.06
59.08
2026.06
58.83
2026.06
58.63
2026.06
58.26
2026.06
57.88
2026.06
45.84
2026.06
38.31