Using real timestamps as rotary position angles in a RoFormer model gives 0.740 AUC on IEEE-CIS fraud detection, about 0.011 higher than the baseline.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Credit Card Fraud Detection Using RoFormer Model With Relative Distance Rotating Encoding
Using real timestamps as rotary position angles in a RoFormer model gives 0.740 AUC on IEEE-CIS fraud detection, about 0.011 higher than the baseline.