Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Topic Classification on AGNEWS (test)
Loading...
56.9
Hit Score (HS)
NA-SFT
5.42
18.785
32.15
45.515
May 28, 2024
Sep 8, 2024
Dec 20, 2024
Apr 2, 2025
Jul 14, 2025
Oct 25, 2025
Feb 5, 2026
Hit Score (HS)
FA Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Hit Score (HS)
FA Score
NA-SFT
Backbone=Mistral-7B
2024.05
56.9
91.5
SFT
Backbone=Mistral-7B
2024.05
52.4
91.9
Vaccine-SFT
Backbone=Mistral-7B
2024.05
48.6
85.7
BSO
Backbone=Mistral-7B
2024.05
47.8
91.2
Vlguard
Backbone=Mistral-7B
2024.05
43.9
90
Lisa
Backbone=Mistral-7B
2024.05
40.5
89.6
SFT
Backbone=Llama3-8B
2026.02
31
90.3
ConstrainedSFT
Backbone=Llama3-8B
2026.02
29.2
84.1
SPARD
Backbone=Llama3-8B
2026.02
24.3
87.9
Lisa
Backbone=Llama3-8B
2026.02
22.7
91
AsFT
Backbone=Llama3-8B
2026.02
18
81.3
SafeGrad
Backbone=Llama3-8B
2026.02
17.5
90.5
DSS
Backbone=Llama3-8B
2026.02
16.9
89.2
Surgery
Backbone=Llama3-8B
2026.02
7.4
88.6
Feedback
Search any
task
Search any
task
