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

Multi-hop Question Answering on 2WikiMHQA

85.56F1 Score

CRAFT7B

-0.562421.796344.15566.5137Jan 25, 2026Feb 15, 2026Mar 9, 2026Mar 30, 2026Apr 21, 2026May 12, 2026Jun 3, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
85.5679.789.9-
2026.02
84.5778.7294.8-
2026.02
81.0170.5595.95-
2026.02
80.6469.9594-
2026.02
79.3170.294.2-
2026.02
79.1870.1594.3-
2026.02
79.0269.6597.15-
2026.02
78.3569.9588.25-
2026.02
77.6768.796.5-
2026.02
72.6363.282.05-
2026.02
71.3262.6578.95-
2026.06
68.7--6,100
2026.06
68.2--10,669
2026.02
67.7359.2578.3-
2026.02
66.7257.776.85-
2026.02
66.2456.390.35-
2026.02
65.6856.0577.2-
2026.06
65.6--6,744
2026.02
64.7556.576.1-
2026.02
64.1653.8586.6-
2026.02
63.1754.778.9-
2026.02
61.3952.0572.25-
2026.06
61.3--4,039
2026.06
59.2--5,130
2026.06
58.7--9,467
2026.02
58.1249.566.45-
2026.06
58--7,753
2026.06
56.3--5,339
2026.04
55.847.7--
2026.04
55.547.7--
2026.04
55.447.9--
2026.06
55.2--6,452
2026.04
55.147.2--
2026.04
54.846.9--
2026.04
54.446.7--
2026.04
53.346.3--
2026.02
52.5143.3557.95-
2026.02
52.445.4538.78-
2026.04
52.244.1--
2026.02
52.0544.747.85-
2026.04
50.744.2--
2026.04
49.342.3--
2026.04
48.640--
2026.04
47.339.6--
2026.04
47.239.8--
2026.04
46.939.9--
2026.06
46.4--7,300
2026.04
4638.4--
2026.04
44.536.8--
2026.04
43.635--
2026.02
41.9734.834.65-
2026.06
41.8--7,160
2026.04
39.431.6--
2026.02
38.830.5545.25-
2026.06
36.7--9,561
2026.02
34.1427.738.05-
2026.02
28.7922.6515.9-
2026.02
28.4622.851.85-
2026.02
21.5315.518.15-
2026.01
19.43---
2026.01
19.3---
2026.01
18.52---
2026.01
17.8---
2026.01
17.56---
2026.02
16.0810.49.1-
2026.01
14.99---
2026.01
14.62---
2026.01
14.61---
2026.01
14.52---
2026.01
14.5---
2026.01
14.46---
2026.01
14.35---
2026.01
14.2---
2026.01
13.89---
2026.01
13.76---
2026.01
13.15---
2026.01
13.02---
2026.01
12.43---
2026.02
9.9361.55-
2026.01
7.91---
2026.01
7.71---
2026.01
7.19---
2026.02
5.131.92.25-
2026.02
4.722.850.68-
2026.02
2.751.051.2-