TCT learns self-supervised embeddings from financial transaction sequences via contrastive coding, achieving AUC 0.8644 alone for fraud detection but showing no improvement when combined with domain-engineered features.
N., Kaiser, Ł., & Polosukhin, I
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Temporal Contrastive Transformer for Financial Crime Detection: Self-Supervised Sequence Embeddings via Predictive Contrastive Coding
TCT learns self-supervised embeddings from financial transaction sequences via contrastive coding, achieving AUC 0.8644 alone for fraud detection but showing no improvement when combined with domain-engineered features.