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

Multi-hop Question Answering on HotpotQA (dev)

81.62Answer F1

SEQGRAPH

17.535234.172650.8167.4474Apr 17, 2020Nov 14, 2020Jun 13, 2021Jan 10, 2022Aug 9, 2022Mar 8, 2023Oct 5, 2023
Updated 1mo ago

Evaluation Results

MethodLinks
2023.07
81.6288.2866.5163.24-
2020.04
81.389.4---
2020.04
79.889---
2023.07
79.39-65.59--
2023.07
7986.8865.3361.52-
2023.07
78.985.765.859.3-
2020.04
78.886.06---
2023.07
77.687.7264.1962.44-
2023.07
75.698662.0360.45-
2023.07
75.2-61.84--
2020.04
75.186.9---
2020.04
74.887---
2020.04
74.786.6---
2020.04
74.384.4---
2020.04
74.386.4---
2020.04
73.386.6---
2020.04
72.285.7---
2020.10
65.8-52.7--
2020.10
62.4-50.2--
2020.10
61.7-50.4--
58.8-46.5--
2023.10
54.7----
2020.10
51.7-42.1--
2023.10
50----
49.4-37.6--
2023.10
48.747---
2023.10
42.336---
2023.10
4248.3---
2023.10
40----
2023.10
39----
2023.10
38.236---
2023.10
36.938.3---
2023.10
36.436---
2023.10
34.732---
2023.10
34.536---
2023.10
34.3----
2023.10
33----
2023.10
31----
2023.10
28.3----
2023.10
27.5----
2023.10
24.7----
2023.10
20.3----
2023.10
20----
2023.04
----39.8
2023.04
----44.6
2023.04
----62.9
2023.04
----61.2
2023.04
----50.2
2023.04
----57
2023.04
----59.2