This paper describes a GCN+BERT fusion model for blockchain fraud detection that reports state-of-the-art F1 scores, but the results are invalid because the label is included in the text input.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Dynamic Feature Fusion: Combining Global Graph Structures and Local Semantics for Blockchain Fraud Detection
This paper describes a GCN+BERT fusion model for blockchain fraud detection that reports state-of-the-art F1 scores, but the results are invalid because the label is included in the text input.