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

Multi-Hop Question Answering on HotpotQA in-domain

46.3EM

SIGHT

-1.85210.64923.1535.651Oct 12, 2025Nov 1, 2025Nov 22, 2025Dec 12, 2025Jan 2, 2026Jan 22, 2026Feb 12, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
46.31.73
2026.02
46.12.18
2025.10
44.6-
2025.10
44.5-
2025.10
44.5-
2025.10
44.5-
2025.10
43.6-
2025.10
43.5-
2025.10
43.3-
2025.10
43.3-
2025.10
42.4-
2026.02
41.991.83
2026.02
41.63.06
2025.10
40.6-
2026.02
40.431.7
2026.02
36.912.95
2026.02
35.742.31
2026.02
35.71.39
2025.10
35.2-
2025.10
33.1-
33.1-
2026.02
32.621.76
2025.10
32.6-
2025.10
30.5-
2025.10
29.9-
2025.10
29.9-
2025.10
28.4-
2025.10
27.1-
2026.02
25.21.19
2025.10
24.2-
2025.10
24.2-
2026.02
23.830
2025.10
23.7-
23.7-
2025.10
21.7-
2025.10
21.7-
2026.02
18.951.91
2025.10
18.7-
2025.10
18.7-
2025.10
18.3-
18.3-
2026.02
17.380
2026.02
14.81
2025.10
13.3-
2025.10
13.3-
2025.10
9.2-
2025.10
9.2-
2026.02
7.420
2026.02
5.661
2026.02
4.10
2026.02
2.543.12
2026.02
00
2026.02
00