Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Robust Smart Contract Vulnerability Detection via Contrastive Learning-Enhanced Granular-ball Training

Zeli Wang, Qingxuan Yang, Shuyin Xia, Yueming Wu, Bo Liu, Longlong Lin

Abstract

Deep neural networks (DNNs) have emerged as a prominent approach for detecting smart contract vulnerabilities, driven by the growing contract datasets and advanced deep learning techniques. However, DNNs typically require large-scale labeled datasets to model the relationships between contract features and vulnerability labels. In practice, the labeling process often depends on existing open-sourced tools, whose accuracy cannot be guaranteed. Consequently, label noise poses a significant challenge for the accuracy and robustness of the smart contract, which is rarely explored in the literature. To this end, we propose Contrastive learning-enhanced Granular-Ball smart Contracts training, CGBC, to enhance the robustness of contract vulnerability detection. Specifically, CGBC first introduces a Granular-ball computing layer between the encoder layer and the classifier layer, to group similar contracts into Granular-Balls (GBs) and generate new coarse-grained representations (i.e., the center and the label of GBs) for them, which can correct noisy labels based on the most correct samples. An inter-GB compactness loss and an intra-GB looseness loss are combined to enhance the effectiveness of clustering. Then, to improve the accuracy of GBs, we pretrain the model through unsupervised contrastive learning supported by our novel semantic-consistent smart contract augmentation method. This procedure can discriminate contracts with different labels by dragging the representation of similar contracts closer, assisting CGBC in clustering. Subsequently, we leverage the symmetric cross-entropy loss function to measure the model quality, which can combat the label noise in gradient computations. Finally, extensive experiments show that the proposed CGBC can significantly improve the robustness and effectiveness of the smart contract vulnerability detection when contrasted with baselines.

Robust Smart Contract Vulnerability Detection via Contrastive Learning-Enhanced Granular-ball Training

Abstract

Deep neural networks (DNNs) have emerged as a prominent approach for detecting smart contract vulnerabilities, driven by the growing contract datasets and advanced deep learning techniques. However, DNNs typically require large-scale labeled datasets to model the relationships between contract features and vulnerability labels. In practice, the labeling process often depends on existing open-sourced tools, whose accuracy cannot be guaranteed. Consequently, label noise poses a significant challenge for the accuracy and robustness of the smart contract, which is rarely explored in the literature. To this end, we propose Contrastive learning-enhanced Granular-Ball smart Contracts training, CGBC, to enhance the robustness of contract vulnerability detection. Specifically, CGBC first introduces a Granular-ball computing layer between the encoder layer and the classifier layer, to group similar contracts into Granular-Balls (GBs) and generate new coarse-grained representations (i.e., the center and the label of GBs) for them, which can correct noisy labels based on the most correct samples. An inter-GB compactness loss and an intra-GB looseness loss are combined to enhance the effectiveness of clustering. Then, to improve the accuracy of GBs, we pretrain the model through unsupervised contrastive learning supported by our novel semantic-consistent smart contract augmentation method. This procedure can discriminate contracts with different labels by dragging the representation of similar contracts closer, assisting CGBC in clustering. Subsequently, we leverage the symmetric cross-entropy loss function to measure the model quality, which can combat the label noise in gradient computations. Finally, extensive experiments show that the proposed CGBC can significantly improve the robustness and effectiveness of the smart contract vulnerability detection when contrasted with baselines.

Paper Structure

This paper contains 20 sections, 13 equations, 4 figures, 5 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Program Analysis Method
  4. Deep Learning Method
  5. LLM-assisted Smart Contract Security Analysis
  6. Robust Method
  7. Comparison with Our Method
  8. Methodology
  9. Contrastive Learning-based Representation Initialization
  10. Smart Contract Augmentation
  11. Contract Representation Pretraining
  12. Robust Vulnerability Detection by Granular-ball Training
  13. Contract Granular-ball Construction
  14. Loss-guided Vulnerability Detection Optimization
  15. Experimental Evaluation
  16. ...and 5 more sections

Figures (4)

  • Figure 1: The illustration of the overall workflow of CGBC, consisting of two main procedures: contrastive learning-based representation initialization and robust vulnerability detection by Granular-ball training.
  • Figure 2: An illustration example of smart contract augmentation.
  • Figure 3: Evaluation results on RQ2: ablation studies on the core modules of CGBC.
  • Figure 4: Evaluation results on RQ4: effectiveness of Graunlar-ball computing in clustering smart contracts across training stages.

Theorems & Definitions (2)

  • Definition 1: Intra-GB Compactness Loss
  • Definition 2: Inter-GB Looseness Loss