Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-class Calibration on 20NG (test)
Loading...
1.4
NLL
NA-FIR
1.164
2.757
4.35
5.943
Dec 9, 2025
NLL
CONF-ECE
BS
CW-ECE
TECE
Updated 4d ago
Evaluation Results
Method
Method
Links
NLL
CONF-ECE
BS
CW-ECE
TECE
NA-FIR
Type=Average Ranking
2025.12
1.4
2.4
2.1
4
3.4
FIR
Type=Average Ranking
2025.12
2.8
3.1
2.8
3.7
4.5
VS
Type=Average Ranking
2025.12
3.2
4.4
4.9
5
5
TS
Type=Average Ranking
2025.12
3.3
6.9
4.9
7
5.6
MS
Type=Average Ranking
2025.12
5.3
4.9
7
3.1
3.7
SCIR
Type=Average Ranking
2025.12
6
2.6
4.3
3.8
3.1
UNCALIBRATED
Type=Average Ranking
2025.12
6.7
7.8
7.3
7.8
7.8
IR OVR
Type=Average Ranking
2025.12
7.3
3.9
2.7
1.6
2.9
Feedback
Search any
task
Search any
task