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

Multi-Hop Question Answering on HotpotQA (EM, F1, SD, SE)

47.1Exact Match (EM)

SD-Search-Instruct

0.312.4524.636.75Apr 19, 2026Apr 23, 2026Apr 28, 2026May 3, 2026May 8, 2026May 13, 2026May 18, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
47.1---
2026.05
46.8---
2026.05
46.8---
2026.05
45.1---
2026.05
44.6---
2026.05
44.5---
2026.05
44.2---
2026.04
43.755.42.6516.5
2026.05
43.1---
2026.04
42.754.21.3431.2
2026.05
42.1---
2026.05
42.1---
2026.05
42---
2026.05
41.4---
2026.05
41---
2026.05
40.8---
2026.05
40.5---
2026.05
40.4---
2026.04
39.450.51.4926.5
2026.04
38.950.42.118.6
2026.04
38.7492.4315.9
2026.05
38.6---
2026.05
38.6---
2026.05
38.6---
2026.05
38.4---
2026.05
38---
2026.05
38---
2026.04
3748.62.0717.8
2026.05
37---
2026.05
36.2---
2026.05
35.1---
2026.05
34.5---
2026.05
33.8---
2026.05
33.4---
2026.04
33.244.22.0516.2
2026.05
33.1---
2026.05
32.9---
2026.05
32.6---
2026.05
32.2---
2026.04
32.142.63.1910.1
2026.04
30.542.71.5519.7
2026.05
30.5---
2026.04
30.339.62.0914.5
2026.05
29.9---
2026.05
29.8---
2026.05
29.7---
2026.04
27.936.71.6317.2
2026.05
27.4---
2026.05
25.5---
2026.05
24.2---
2026.05
23.7---
2026.05
22.8---
2026.05
22.1---
2026.05
21.7---
2026.05
20.8---
2026.05
20.1---
2026.05
18.7---
2026.05
18.6---
2026.05
18.3---
2026.05
16.4---
2026.05
14.9---
2026.05
13.3---
2026.04
10.319.11.666.22
2026.04
10.118.41.586.4
2026.05
9.2---
2026.05
2.1---