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

Multi-Hop Question Answering on HotpotQA 2018 Wikipedia dump (dev)

46.8Accuracy

MR-Search

13.62422.23730.8539.463Mar 11, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
46.8
2026.03
43.9
2026.03
43.9
2026.03
41.9
2026.03
38.6
2026.03
37.8
2026.03
37.3
2026.03
35.3
2026.03
32.6
2026.03
30.5
2026.03
22.1
2026.03
18.7
2026.03
18.3
2026.03
14.9