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

Multi-Hop Question Answering on MuSiQue (EM)

25.3Exact Match (EM)

EvalAct

-0.8045.97312.7519.527Feb 3, 2026Feb 13, 2026Feb 24, 2026Mar 7, 2026Mar 17, 2026Mar 28, 2026Apr 8, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.03
25.3
2026.03
21.6
2026.03
18.7
2026.04
18.1
2026.04
17.4
2026.04
17.2
2026.04
16
2026.03
15.7
2026.03
14.6
2026.02
12.5
2026.04
12.4
2026.04
12.4
2026.02
12
2026.04
11.9
2026.02
11.1
2026.03
10.3
2026.04
10.1
2026.04
9.9
2026.04
9.8
2026.02
9.1
2026.02
9.1
2026.02
8.9
2026.02
8.3
2026.02
7.6
2026.02
7.4
2026.02
7.2
2026.02
7.2
2026.03
7.2
2026.02
6.9
2026.02
6.8
2026.02
6.8
2026.02
6.7
2026.03
6.7
2026.04
6.6
2026.02
6.4
2026.02
6.3
2026.02
6.3
2026.03
5.8
2026.03
5.8
2026.02
5.5
2026.04
5.5
2026.03
5.4
2026.04
5.4
2026.02
4.9
2026.02
4.7
2026.02
4.7
2026.03
4.7
2026.04
4.7
2026.04
4.4
2026.02
3.1
2026.03
3.1
2026.02
2.7
2026.02
2.2
2026.02
2
2026.03
2
2026.04
2
2026.02
0.2
2026.04
0.2