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

Multi-Hop Question Answering on 2Wiki (out-of-domain)

47.6Accuracy

ReSearch-7B-Instruct

9.6419.49529.3539.205Oct 15, 2025Nov 3, 2025Nov 23, 2025Dec 12, 2025Jan 1, 2026Jan 20, 2026Feb 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
47.6
2025.10
44.7
2025.10
42.3
2026.02
42
2025.10
42
2025.10
41.9
2025.10
41.2
2026.02
40.3
2026.02
40.1
2025.10
38.2
2025.10
37.4
2026.02
36.6
2025.10
36.3
2026.02
35.2
2025.10
29.6
2025.10
29.2
2026.02
27.5
2025.10
27.3
2026.02
27
2025.10
25.9
2025.10
25
2025.10
23.5
2026.02
23.2
2026.02
22.6
2026.02
22.2
2025.10
17.6
2025.10
14.9
2025.10
11.1