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

Multi-hop QA on Bamboogle (Accuracy %)

74.9Accuracy (%)

IGPO

20.09234.32148.5562.779Mar 18, 2026Mar 20, 2026Mar 23, 2026Mar 25, 2026Mar 28, 2026Mar 30, 2026Apr 2, 2026
Updated 15d ago

Evaluation Results

MethodLinks
2026.03
74.9
2026.04
73.87
72.8
2026.04
70.83
2026.03
68.9
2026.04
68.9
2026.04
65.87
2026.04
64.27
2026.04
62.13
2026.04
50.93
2026.03
42
2026.03
40.3
2026.03
40.1
2026.03
35.2
2026.03
33.2
2026.03
27.5
2026.03
27
2026.03
26.2
2026.03
25.4
2026.03
25
2026.03
24.8
2026.03
23.7
2026.03
23.2
2026.03
22.6
2026.03
22.2