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

Multi-hop Question Answering on Bamboogle

75.2Accuracy

TC+FM*

1.15220.37639.658.824Oct 9, 2023Mar 3, 2024Jul 28, 2024Dec 22, 2024May 17, 2025Oct 11, 2025Mar 7, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
75.2
2026.02
70.4
2026.03
69
2026.03
68.9
2026.02
68.8
2026.02
67.2
2026.03
64.5
2026.03
64.1
2026.02
63.2
2026.02
60.8
2026.02
60
2026.02
60
2026.02
59.2
2026.02
58.4
2026.02
58.4
2026.02
58.4
2026.02
54.4
2026.02
52.8
2026.02
51.2
2023.10
50.4
2026.02
50.4
2026.02
48.8
2026.02
48
2026.02
47.2
2026.02
46.4
2026.02
44.8
2026.02
44.8
2026.02
44.8
2026.02
44.8
2023.10
44
2026.02
43.2
2026.02
43.2
2023.10
43.2
2026.02
43
2026.02
42.4
2023.10
41.6
2023.10
40.8
2026.02
40.6
2026.02
39.2
2026.02
39.2
2026.02
38.4
2026.03
36.8
2026.02
36
2026.02
36
2026.02
34.4
2026.02
34.4
2026.02
31.2
2026.02
29.6
2026.03
29.3
2026.03
27.8
2026.02
27.2
2026.02
26.6
2026.02
26.4
2026.03
26.4
2026.02
22.4
2026.02
22.4
2026.02
20.8
2026.03
19.2
2026.02
11.2
2026.03
11.1
2023.10
7.2
2026.02
4