Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Hallucination prediction on ParaRel + domain
Loading...
69.24
Accuracy
Conf + Probe (SCAO)
56.812
60.0385
63.265
66.4915
Sep 18, 2025
Accuracy
A(phi(sM))
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
A(phi(sM))
Conf + Probe (SCAO)
Model Size=8B
2025.09
69.24
14.15
Conf (SCAO)
Model Size=70B
2025.09
69.19
-
Conf + Probe
Model Size=8B
2025.09
68.65
13.56
Conf (SCAO)
Model Size=8B
2025.09
67.88
-
Probe_dnn
Model Size=8B
2025.09
66.82
11.73
Conf + Probe (SCAO)
Model Size=70B
2025.09
66.36
13.14
Probe_dnn
Model Size=70B
2025.09
65.24
12.02
Conf + Probe
Model Size=70B
2025.09
64.46
11.24
Conf
Model Size=8B
2025.09
62.63
-
Conf
Model Size=70B
2025.09
57.29
-
Feedback
Search any
task
Search any
task