Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Natural Language Inference on MedNLI
Loading...
100
AR (%)
Fixed-Threshold
26.056
45.253
64.45
83.647
May 18, 2026
AR (%)
Risk (%)
Updated 13d ago
Evaluation Results
Method
Method
Links
AR (%)
Risk (%)
Fixed-Threshold
alpha=0.20
2026.05
100
21
Always-Act
alpha=0.20
2026.05
100
21
Naive-Tuning
alpha=0.20
2026.05
99
20.8
ACI
alpha=0.20
2026.05
97.2
20.3
SAOCP
alpha=0.20
2026.05
96
20.2
NEX-Conf
alpha=0.20
2026.05
92.5
18.3
ConfFact
alpha=0.20
2026.05
76.8
13.5
LTT
alpha=0.20
2026.05
66.6
11.4
CSA
alpha=0.20
2026.05
28.9
6.7
CRC
alpha=0.20
2026.05
28.9
4.9
Feedback
Search any
task
Search any
task