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

Accuracy on Multi-Hop Question Answering on Bamboogle (test)

70.1Accuracy

GRPO w/ PaW

8.7424.6740.656.53Jun 1, 2026Jun 6, 2026Jun 12, 2026Jun 18, 2026Jun 24, 2026Jun 30, 2026Jul 6, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.06
70.1
2026.06
69.6
2026.06
69.5
2026.07
69.4
2026.06
68.9
2026.07
68.9
2026.06
65.2
2026.06
64.9
2026.07
64.9
2026.06
64.1
2026.07
64.1
2026.06
60
2026.07
40
2026.06
36.8
2026.07
36.8
2026.07
34.4
2026.07
29.3
2026.06
27.8
2026.07
27.8
2026.06
26.4
2026.07
26.4
2026.07
19.2
2026.06
11.1
2026.07
11.1