Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Natural Language Explanation Generation on e-SNLI
Loading...
50
Human Evaluation Score
AdaLora
15.9712
24.8056
33.64
42.4744
May 22, 2023
Human Evaluation Score
IAA (Cohen's Kappa)
Quality Score
Error Rate
Data Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Human Evaluation Score
IAA (Cohen's Kappa)
Quality Score
Error Rate
Data Score
AdaLora
Backbone=Llama-2-7B
2023.05
50
-
-
-
-
SPARSEFIT
Backbone=Llama-2-7B, S...
2023.05
41.11
-
-
-
-
SPARSEFIT (Att.Q+LN)
Backbone=T5-large, Fin...
2023.05
38.27
0.34
-
-
-
Full Fine-tuning
Backbone=T5-large, Fin...
2023.05
29.63
0.43
-
-
-
AdaLora
Backbone=T5-large, Fin...
2023.05
23.33
0.34
-
-
-
Full Fine-tuning
Backbone=Llama-2-7B
2023.05
17.78
-
-
-
-
SPARSEFIT (Att.Q)
Backbone=T5-large, Fin...
2023.05
17.28
0.38
-
-
-
Direct Generation
Approach=Direct Genera...
2026.04
-
-
62
28
100
Generate+Rerank
Approach=Generate+Rerank
2026.04
-
-
76
15
60
PPO+Ranking
Approach=PPO+Ranking
2026.04
-
-
82
9
40
Feedback
Search any
task
Search any
task