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

Question Answering on HotpotQA (dev)

81Answer F1

COS

7.57626.63845.764.762Jul 28, 2020Feb 7, 2021Aug 21, 2021Mar 3, 2022Sep 14, 2022Mar 27, 2023Oct 8, 2023
Updated 1mo ago

Evaluation Results

MethodLinks
2023.05
8185.372.368.261.146.4
2023.05
80.986.572.568.161.545.9
2023.05
80.184.571.467.360.245.3
2023.05
79.281.86966.65642
2023.05
76.979.1-62.951.3-
2020.07
75.786.867.7---
2020.07
75.587.167.8---
2023.05
75.179.466.362.356.542.1
2020.07
74.384.464.4---
2020.07
73.583.463.5---
2023.05
73.376.161.460.549.235.8
2023.05
66.272.152.95441.727.7
2023.10
65.9-----
2023.10
64.9-----
2023.10
63.1-----
2023.10
61.7-----
2023.10
61.6-----
2023.10
61.6-----
2023.10
61.1-----
2023.10
59.7-----
2023.05
58.871.549.246.539.926.6
2023.10
57.1-----
2023.10
56.3-----
2023.10
54.7-----
2023.10
54.5-----
2023.10
52.9-----
2023.05
49.458.535.337.623.112.2
2023.10
46.1-----
2023.10
41.4-----
2023.10
41.3-----
2023.05
40.447.727.631.11711.8
2023.10
39.8-----
2023.10
36.6-----
2023.10
34.7-----
2023.10
34-----
2023.10
27.3-----
2023.10
24-----
2023.10
23.3-----
2023.10
15-----
2023.10
13.2-----
2023.10
12.5-----
2023.10
10.5-----
2023.10
10.4-----
2020.12
-89.0773.12---
2020.12
-89.2173.57---