Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Hallucination Correction on Hallucination
Loading...
15.15
Error Rate (ERR)
SoLA
-311.592
1,893.9165
4,099.425
6,304.9335
Mar 11, 2026
Error Rate (ERR)
True Recall Rate (TRR)
Adjusted Recall Rate (ARR)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Error Rate (ERR)
True Recall Rate (TRR)
Adjusted Recall Rate (ARR)
SoLA
Backbone=GPT2-XL
2026.03
15.15
1.01
7.35
GRACE
Backbone=GPT2-XL
2026.03
15.84
7.14
10
ELDER
Backbone=GPT2-XL
2026.03
16.12
5.87
8.42
MELO
Backbone=GPT2-XL
2026.03
17.45
1.04
2.66
ROME
Backbone=GPT2-XL
2026.03
30.28
103.82
14.02
MEND
Backbone=GPT2-XL
2026.03
1,369.8
1,754.9
2,902.5
CMR
Backbone=GPT2-XL
2026.03
1,449.3
28.14
107.76
EWC
Backbone=GPT2-XL
2026.03
1,485.7
29.24
109.59
CLEAR
Backbone=GPT2-XL
2026.03
2,394.3
35.34
195.82
SERAC
Backbone=GPT2-XL
2026.03
8,183.7
133.3
10.04
Feedback
Search any
task
Search any
task