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

Multi-hop Question Answering on 2Wiki (test)

69.7F1 Score

CatRAG

8.75624.57840.456.222Dec 23, 2025Jan 7, 2026Jan 23, 2026Feb 8, 2026Feb 23, 2026Mar 11, 2026Mar 27, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
69.7---
2026.02
68.1---
2026.02
56.9---
2026.02
56.7---
2026.02
52.8---
2026.02
49.7---
2026.03
49.33--41.52
2026.03
48.66--40.6
2026.03
43.54--35.27
2026.02
42.8---
2025.12
42.7---
2026.02
41.9---
2025.12
41.4---
2026.02
39.9---
2026.03
39.03--33.08
2026.03
37.87--32.11
2026.03
34.17--27.85
2026.03
32.22--26.82
2026.03
32.01--25.59
2026.03
31.47--25.97
2026.03
30.32--25.52
2026.03
30.18--24.54
2026.03
30--21.45
2025.12
29.6---
2026.03
29.28--21.49
2025.12
29.2---
2026.03
28.73--24.79
2025.12
27.3---
2025.12
25.9---
2025.12
25---
2025.12
23.5---
2025.12
17.6---
2025.12
14.9---
2025.12
11.1---
2025.02
-0.5495--
2025.02
-0.585--
2025.02
-0.5971--
2025.02
-0.6314--
2025.02
-0.4902--
2025.02
-0.5117--
2025.02
-0.4825--
2025.02
-0.5218--