KGBERT4Eth: A Feature-Complete Transformer Powered by Knowledge Graph for Multi-Task Ethereum Fraud Detection

• URL: http://arxiv.org/abs/2509.03860v1
• Date: Thu, 04 Sep 2025 03:38:11 GMT
• Title: KGBERT4Eth: A Feature-Complete Transformer Powered by Knowledge Graph for Multi-Task Ethereum Fraud Detection
• Authors: Yifan Jia, Ye Tian, Liguo Zhang, Yanbin Wang, Jianguo Sun, Liangliang Song,
• Abstract summary: KGBERT4Eth is a feature-complete pre-training encoder that combines Transaction Semantics and Transaction Knowledge Graph.<n>We optimize pre-training objectives for both components to fuse these complementary features, generating feature-complete embeddings.<n> KGBERT4Eth significantly outperforms state-of-the-art baselines in both phishing account detection and de-anonymization tasks.
• Score: 14.186303004456205
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Ethereum's rapid ecosystem expansion and transaction anonymity have triggered a surge in malicious activity. Detection mechanisms currently bifurcate into three technical strands: expert-defined features, graph embeddings, and sequential transaction patterns, collectively spanning the complete feature sets of Ethereum's native data layer. Yet the absence of cross-paradigm integration mechanisms forces practitioners to choose between sacrificing sequential context awareness, structured fund-flow patterns, or human-curated feature insights in their solutions. To bridge this gap, we propose KGBERT4Eth, a feature-complete pre-training encoder that synergistically combines two key components: (1) a Transaction Semantic Extractor, where we train an enhanced Transaction Language Model (TLM) to learn contextual semantic representations from conceptualized transaction records, and (2) a Transaction Knowledge Graph (TKG) that incorporates expert-curated domain knowledge into graph node embeddings to capture fund flow patterns and human-curated feature insights. We jointly optimize pre-training objectives for both components to fuse these complementary features, generating feature-complete embeddings. To emphasize rare anomalous transactions, we design a biased masking prediction task for TLM to focus on statistical outliers, while the Transaction TKG employs link prediction to learn latent transaction relationships and aggregate knowledge. Furthermore, we propose a mask-invariant attention coordination module to ensure stable dynamic information exchange between TLM and TKG during pre-training. KGBERT4Eth significantly outperforms state-of-the-art baselines in both phishing account detection and de-anonymization tasks, achieving absolute F1-score improvements of 8-16% on three phishing detection benchmarks and 6-26% on four de-anonymization datasets.

Related papers

• LMAE4Eth: Generalizable and Robust Ethereum Fraud Detection by Exploring Transaction Semantics and Masked Graph Embedding [10.923718297125754]
  LMAE4Eth is a multi-view learning framework that fuses transaction semantics, masked graph embedding, and expert knowledge.<n>We first propose a transaction-token contrastive language model (TxCLM) that transforms context-independent numerical transaction records into cohesive linguistic representations.<n>We then propose a masked account graph autoencoder (MAGAE) using generative self-supervised learning, which achieves superior node-level account detection.
  arXiv  Detail & Related papers  (2025-09-04T06:56:32Z)
• Correlating Account on Ethereum Mixing Service via Domain-Invariant feature learning [13.405407918706254]
  Untraceability of transactions facilitated by mixing services like Tornado Cash poses significant challenges to blockchain security and financial regulation.<n>We propose StealthLink, a novel framework that addresses these limitations through cross-task domain-invariant feature learning.<n>Experiments on real-world mixing transaction datasets demonstrate that StealthLink achieves state-of-the-art performance, with 96.98% F1-score in 10-shot learning scenarios.
  arXiv  Detail & Related papers  (2025-05-15T01:27:12Z)
• Dynamic Feature Fusion: Combining Global Graph Structures and Local Semantics for Blockchain Fraud Detection [0.7510165488300369]
  We propose a dynamic feature fusion model that combines graph-based representation learning and semantic feature extraction for fraud detection.<n>We develop a comprehensive data processing pipeline, including graph construction, temporal feature enhancement, and text preprocessing.<n> Experimental results on large-scale real-world blockchain datasets demonstrate that our method outperforms existing benchmarks across accuracy, F1 score, and recall metrics.
  arXiv  Detail & Related papers  (2025-01-03T09:04:43Z)
• Ethereum Fraud Detection via Joint Transaction Language Model and Graph Representation Learning [6.378807038086552]
  Current fraud detection methods fail to consider the semantic information and similarity patterns within transactions.<n>We propose TLMG4Eth that combines a transaction language model with graph-based methods to capture semantic, similarity, and structural features of transaction data.
  arXiv  Detail & Related papers  (2024-09-09T07:13:44Z)
• Facilitating Feature and Topology Lightweighting: An Ethereum Transaction Graph Compression Method for Malicious Account Detection [3.877894934465948]
  Bitcoin has become one of the primary global platforms for cryptocurrency, playing an important role in promoting the diversification of the financial ecosystem. Previous regulatory methods usually detect malicious accounts through feature engineering or large-scale transaction graph mining. We propose an Transaction Graph Compression method named TGC4Eth, which assists malicious detection by lightweighting both features and topology of the transaction graph.
  arXiv  Detail & Related papers  (2024-05-14T02:21:20Z)
• UGMAE: A Unified Framework for Graph Masked Autoencoders [67.75493040186859]
  We propose UGMAE, a unified framework for graph masked autoencoders. We first develop an adaptive feature mask generator to account for the unique significance of nodes. We then design a ranking-based structure reconstruction objective joint with feature reconstruction to capture holistic graph information.
  arXiv  Detail & Related papers  (2024-02-12T19:39:26Z)
• Cross-BERT for Point Cloud Pretraining [61.762046503448936]
  We propose a new cross-modal BERT-style self-supervised learning paradigm, called Cross-BERT. To facilitate pretraining for irregular and sparse point clouds, we design two self-supervised tasks to boost cross-modal interaction. Our work highlights the effectiveness of leveraging cross-modal 2D knowledge to strengthen 3D point cloud representation and the transferable capability of BERT across modalities.
  arXiv  Detail & Related papers  (2023-12-08T08:18:12Z)
• Cross-modal Orthogonal High-rank Augmentation for RGB-Event Transformer-trackers [58.802352477207094]
  We explore the great potential of a pre-trained vision Transformer (ViT) to bridge the vast distribution gap between two modalities. We propose a mask modeling strategy that randomly masks a specific modality of some tokens to enforce the interaction between tokens from different modalities interacting proactively. Experiments demonstrate that our plug-and-play training augmentation techniques can significantly boost state-of-the-art one-stream and two trackersstream to a large extent in terms of both tracking precision and success rate.
  arXiv  Detail & Related papers  (2023-07-09T08:58:47Z)
• Object Segmentation by Mining Cross-Modal Semantics [68.88086621181628]
  We propose a novel approach by mining the Cross-Modal Semantics to guide the fusion and decoding of multimodal features. Specifically, we propose a novel network, termed XMSNet, consisting of (1) all-round attentive fusion (AF), (2) coarse-to-fine decoder (CFD), and (3) cross-layer self-supervision.
  arXiv  Detail & Related papers  (2023-05-17T14:30:11Z)
• Entity-Graph Enhanced Cross-Modal Pretraining for Instance-level Product Retrieval [152.3504607706575]
  This research aims to conduct weakly-supervised multi-modal instance-level product retrieval for fine-grained product categories. We first contribute the Product1M datasets, and define two real practical instance-level retrieval tasks. We exploit to train a more effective cross-modal model which is adaptively capable of incorporating key concept information from the multi-modal data.
  arXiv  Detail & Related papers  (2022-06-17T15:40:45Z)
• Cross-Supervised Joint-Event-Extraction with Heterogeneous Information Networks [61.950353376870154]
  Joint-event-extraction is a sequence-to-sequence labeling task with a tag set composed of tags of triggers and entities. We propose a Cross-Supervised Mechanism (CSM) to alternately supervise the extraction of triggers or entities. Our approach outperforms the state-of-the-art methods in both entity and trigger extraction.
  arXiv  Detail & Related papers  (2020-10-13T11:51:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.