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

Multi-Hop Question Answering on HotpotQA (in-domain) (Accuracy)

45.1Accuracy

GEPO

15.04422.84730.6538.453Oct 30, 2025Nov 16, 2025Dec 3, 2025Dec 20, 2025Jan 6, 2026Jan 23, 2026Feb 9, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2025.10
45.1
2026.02
43.2
2025.10
41.6
2025.10
40.6
2026.02
38.6
2025.10
38.6
2026.02
38.2
2026.02
37
2025.10
37
2025.10
36.9
2026.02
34.6
2025.10
34.6
2025.10
34.5
2025.10
32.4
2025.10
27.4
2026.02
25.8
2025.10
23.7
2026.02
20.8
2025.10
20.8
2026.02
17
2026.02
16.4
2026.02
16.2