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

Multi-Hop Question Answering on 2Wiki (token-level F1)

58.7Token-Level F1

COSEARCH

10.96423.35735.7548.143Apr 19, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
58.7
2026.04
58.4
2026.04
56
2026.04
55.9
2026.04
55.6
2026.04
54.7
2026.04
51
2026.04
42.8
2026.04
41.1
2026.04
40.6
2026.04
39.7
2026.04
38.7
2026.04
30.4
2026.04
28.6
2026.04
26.4
2026.04
24.9
2026.04
24.4
2026.04
22.9
2026.04
19.3
2026.04
12.8