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

Question Answering on 2WikiMultihopQA

62.5Accuracy

GRPO

24.12434.08744.0554.013Jun 13, 2024Sep 20, 2024Dec 29, 2024Apr 8, 2025Jul 17, 2025Oct 25, 2025Feb 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
62.5-
2026.02
61.7-
2026.02
61.1-
2026.02
60.7-
2026.02
59.3-
2026.02
57.5-
2026.02
57-
2026.02
56.2-
2026.02
55.17-
2026.02
54.3-
2026.02
54.2-
2026.02
38.5-
2024.06
36.31,097.1
2024.06
31.746.3
2024.06
3144.2
2024.06
30.360.4
2024.06
29.735.7
2024.06
2966.5
2024.06
28.6847.6
2024.06
28.42,356.9
2024.06
28.367
2024.06
27.635.5
2024.06
26.745.1
2024.06
26.335.7
2026.02
25.6-