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

Knowledge-Intensive Reasoning on 2WikiMultiHopQA

48.8Accuracy

AutoTool (Qwen3-8B)

8.86419.23229.639.968Dec 15, 2025Dec 22, 2025Dec 30, 2025Jan 7, 2026Jan 14, 2026Jan 22, 2026Jan 30, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
48.8
2025.12
44.5
2025.12
36.5
2025.12
29.8
2026.01
29.5
2026.01
29.5
2026.01
28.5
2026.01
26.5
2026.01
25
2025.12
22.3
2026.01
21.5
2026.01
20.5
2025.12
19.7
2026.01
17
16.7
14.1
2026.01
14
10.4