Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-Session Dialogue Generation on MSC Human Evaluation 1.0 (test)
Loading...
24.2
Reference Own Topic
SumMem-MSC 2.7B
14.632
17.116
19.6
22.084
Jul 15, 2021
Reference Own Topic
Reference Other's Topic
New Topic
Engaging Response
Final Rating
Updated 4d ago
Evaluation Results
Method
Method
Links
Reference Own Topic
Reference Other's Topic
New Topic
Engaging Response
Final Rating
SumMem-MSC 2.7B
Scale=2.7B, Retrieval...
2021.07
24.2
26.4
78.3
59.3
3.68
SumMem-MSC 2.7B
Scale=2.7B, Retrieval...
2021.07
22.1
30.7
76.4
58.9
3.62
BST 2.7B
Scale=2.7B
2021.07
19.9
14.5
69
53
3.14
SumMem-MSC 2.7B
Scale=2.7B, Retrieval...
2021.07
19.6
33.8
72.7
62.1
3.65
MSC 2.7B
Scale=2.7B, Truncation...
2021.07
15.8
21.8
75.8
56.5
3.29
MSC 2.7B
Scale=2.7B, Truncation...
2021.07
15
22.5
74.4
54.2
3.47
Feedback
Search any
task
Search any
task