Pith. sign in

REVIEW 5 major objections 6 minor 94 references

CASPER: Contrastive Approach for Smart Ponzi Scheme Detecter with More Negative Samples

T0 review · 5 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read CASPER shows that contrastive pre-training on unlabeled smart-contract source code detects Ponzi schemes at F1 92.4 with only 25% of the labels, beating a fully supervised baseline.

desk verdict The label-efficiency gains are almost certainly an artifact of an augmentation that deletes the Ponzi-defining logic; the paper has a plausible idea but the central empirical claim is not supported by the described pipeline. read the letter →

arxiv 2507.16840 v1 pith:P5LCM6BW submitted 2025-07-19 cs.CR cs.AI

Weijia Yang , Tian Lan , Leyuan Liu , Wei Chen , Tianqing Zhu , Sheng Wen , Xiaosong Zhang This is my paper · ORCID
classification cs.CRcs.AI
keywords smartPonzischemedetectioncontrastivelearningsemi-supervisedclassificationcontractsecurityEthereumpseudo-labelself-trainingmulti-vectorcosinesimilarityGraphCodeBERT
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper is trying to show that smart Ponzi scheme detection does not have to be a fully supervised problem. It proposes CASPER, which first learns representations of smart-contract source code from 10,051 unlabeled contracts using a three-view contrastive objective, then trains a semi-supervised classifier that expands its own training set with confident pseudo-labels. On the XBlock dataset, CASPER reaches F1 92.4 with only 25% of the labels, against 79.3 for the SourceP baseline under identical conditions, and 95.2 with full labels. The same pre-trained encoder transfers to other Ponzi datasets and to honeypot and phishing contracts. If these numbers hold, the main bottleneck of scarce labels can be pushed off the critical path.

What carries the argument

The central object is a three-view contrastive objective over augmented smart-contract source codes. Each contract is transformed by three augmentations: splitting variables into two to five sub-variables, replacing function bodies with a simple return statement, and renaming variables. Each view's source code is parsed into an abstract syntax tree and a data flow graph, and GraphCodeBERT takes the source code plus the data flow graph as input to produce a feature vector. The contrastive loss maximizes a multi-vector cosine similarity among the three views of the same contract while minimizing similarity against the $N^3 - N$ negative pairs formed by views of different contracts in the batch, with the equiangular intermediate vector representing the similarity of each group. A self-training classifier then assigns pseudo-labels to confident unlabeled predictions and retrains iteratively.

What would settle it

A direct check: train the encoder on the 10,051 unlabeled contracts with only the medium augmentation (function bodies replaced by a return statement), then see whether a linear probe on the resulting representations can still separate the 318 known Ponzi contracts from the non-Ponzi contracts in XBlock; if it cannot, the contrastive objective has learned away the body-level payout logic that distinguishes Ponzi contracts. A second, cheaper check: search for address or source-code hash overlaps between the 10,051 pre-training contracts and the XBlock, EPSD, EBD, honeypot, and phishing sets—if overlap exists, the label-efficiency advantage may be partially inherited rather than learned.

Watch

Extended reading notes

Core claim

The central claim is that a contrastive representation-learning stage over unlabeled code, followed by self-trained semi-supervised classification, yields smart Ponzi detection that is both more accurate and far less label-hungry than fully supervised source-code analysis. The authors compare CASPER against SourceP with the same label settings: with 100% labels CASPER's F1 is 95.2 versus 90.3, and with 25% labels it is 92.4 versus 79.3. They further report that the learned encoder generalizes to the EPSD and EBD Ponzi datasets, to unseen Ponzi scheme types, and to unrelated fraud types such as honeypot and phishing contracts, which they read as evidence that the representation captures general contract semantics rather than a memorized dataset. The paper also contributes a multi-vector cosine similarity measure, built on an intermediate vector that makes equal angles with all input vectors, and reports that it converges faster than centroid or weighted-average similarity.

Load-bearing premise

The load-bearing premise is that the three code augmentations—splitting variables, replacing function bodies with a return statement, and renaming variables—preserve the semantic signals that mark a contract as a Ponzi scheme; if an augmentation discards payout or withdrawal logic, the contrastive encoder will learn representations blind to the very cues the classifier needs, and the 10,051-contract pre-training corpus must also be assumed not to overlap the evaluation sets.

Editorial extensions

If this is right

  • Smart Ponzi detection can be built with far fewer hand-labeled contracts: with 25% labels CASPER already exceeds the F1 of SourceP trained on 100% labels.
  • Unlabeled source code from public blockchain explorers can carry the representation-learning burden, reducing annotation cost and making large-scale screening more practical.
  • The pre-trained encoder transfers to other Ponzi datasets and to other fraud types, so one contrastive pre-training run may serve multiple detection tasks.
  • The learned representation is classifier-agnostic: SVM, XGBoost, and a multilayer perceptron all reach F1 scores above 93, indicating that the representation, not the final classifier, drives the gain.
  • Expanding the negative-sample pool from $2N-1$ (SimCLR) and $N^2-N$ (CLIP) to $N^3-N$ improves downstream classification in the reported comparison.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable consequence the paper does not report: if the 10,051-contract pre-training corpus has no overlap with the evaluation sets, then the same three-view contrastive recipe could be dropped onto other blockchain fraud labels without architectural changes.
  • The augmentations' semantics deserve scrutiny: replacing function bodies with a return statement could teach the encoder to ignore the payout logic that distinguishes Ponzi contracts, so a synthetic check of whether augmented views retain Ponzi-relevant cues would settle whether the label-efficiency gain is robust to augmentation choice.
  • The paper attributes part of its gain to the multi-vector cosine similarity, but it does not isolate the equiangular intermediate vector from the larger negative pool; ablating the intermediate vector while keeping the $N^3-N$ negative pairs would clarify which component matters.
  • Because CASPER works from static source code and data flow graphs, it can in principle screen contracts before deployment, unlike transaction-based detectors that only react after funds have moved.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper proposes CASPER, a semi-supervised framework for detecting smart Ponzi schemes in Ethereum contracts. The method first pre-trains a GraphCodeBERT encoder with a contrastive objective on unlabeled contract source code, using three code augmentations (variable splitting, function-body replacement, and variable renaming) to generate views, and a three-view similarity measure based on an 'intermediate vector' with equal angles to the three view representations. A classifier is then trained with labeled plus pseudo-labeled data via confidence-thresholded self-training. The authors report that on the XBlock dataset, CASPER reaches F1 92.4 with only 25% labels and F1 95.2 with 100% labels, outperforming SourceP and several other baselines, and they further report generalization and transfer results on EPSD, EBD, honeypot, and phishing datasets. The paper also contains a proof and derivation for the multi-vector cosine similarity method.

Significance. If the reported results are correct, CASPER would be a practically valuable contribution: it would show that self-supervised pre-training on unlabeled source code plus pseudo-label self-training can substantially reduce the label cost of smart Ponzi detection while improving accuracy over fully supervised source-code baselines. The proposed three-view contrastive objective with an O(N^3) negative sample pool is conceptually interesting, and the authors provide a new similarity formulation. However, the significance is heavily conditional: the central experimental claims are undermined by internal inconsistencies (abstract vs. table, text vs. table), a lack of error bars, and, most importantly, the described medium augmentation destroys the function-body semantics that the paper's own ablation suggests are crucial. These issues cast doubt on whether the reported performance can be attributed to the described method.

major comments (5)
  1. [Section III-B (Eq. 5–6) and Section V-B (Fig. 5)] The medium augmentation replaces every function body with 'return value;' and the contrastive loss (Eq. 12) trains the encoder to be invariant to this replacement. This removes exactly the payout, withdrawal, and transfer logic that distinguishes a Ponzi contract from a legitimate one, and the DFG of the stripped view is essentially edge-free. The paper's own ablation in Figure 5 shows that DFG (i.e., data-flow context between functions) is the more important input. As described, the method should therefore discard the most discriminative features; the reported F1 of 92.4 at 25% labels is not credible unless an unstated compensating factor exists. Please run a control experiment with a semantics-preserving medium augmentation (e.g., dead-code insertion or statement reordering) and report whether all three views remain necessary; also probe the learned encoder with a linear classifier on function-body-level semantics to verify that the augmented views retain Ponzi-relevant information.
  2. [Section V-B.1, Table III vs. abstract] The headline numbers in the abstract do not match the results in Table III. The abstract states that with 100% labels CASPER outperforms the baseline by 2.3% in F1 score, but Table III shows a difference of 4.9 percentage points (95.2 vs. 90.3); with 25% labels the abstract claims 'nearly 20% higher,' but Table III shows a difference of 13.1 points (92.4 vs. 79.3). The text in Section V-B.1 also says SourceP's F1 is higher at 25% than at 50%, yet Table III lists 79.3 at 25% and 84.2 at 50%. These contradictions must be resolved before the reported performance can be assessed.
  3. [Section V-A.1 and Section V-B.1] The self-supervised pre-training corpus (10,051 contracts crawled from Etherscan) and the XBlock evaluation set (6,498 contracts from Etherscan) are both built from Etherscan, but no overlap check is reported. If any evaluation contract appears in the pre-training corpus, the reported label-efficiency gains could be partly due to test-time memorization rather than representation quality. The same concern applies to the EPSD, EBD, honeypot, and phishing evaluation sets. Please report a hash-based or address-based overlap analysis between the pre-training corpus and each evaluation set, and re-run the experiments after removing any overlapping contracts.
  4. [Section V-B.1] The paper explicitly states that the block-height-based split 'can provide better model performance' than a random division. Choosing the split strategy that maximizes performance for the main comparison, without reporting the random-split result or any variance across seeds, makes the headline F1 values non-representative. Report mean and standard deviation over at least five random seeds for both the temporal split and a random split, for both CASPER and SourceP, and state which split was used for each table.
  5. [Section IV-A and IV-B] The proof of the existence of an equal-angle intermediate vector is invalid. In Section IV-A, the plane H perpendicular to an arbitrary vector v need not intersect all three given vectors, and the circumcenter argument equates distances from a line to points A, B, C, not the angles between the direction vector v and the original vectors. The subsequent derivation is restricted to R^3 and contains typographical errors (Eq. 21 repeats 'αxa' in all three terms; Eq. 19's third denominator appears to use z_b instead of z_c). Since the actual feature vectors produced by GraphCodeBERT are 768-dimensional, the paper must provide a correct existence and construction argument for R^d, or state explicitly which approximation is used in higher dimensions.
minor comments (6)
  1. [Title/Abstract] 'Detecter' is a typo; also the abstract says 'smart Ponzi scheme detectER' while the title uses the same misspelling.
  2. [Section III-B] The naming of augmentations is inconsistent: the text calls the function-body-replacement augmentation 'medium,' the renaming augmentation 'weak,' but then uses subscripts w (weak) for the function-body view and m (medium) for the renaming view. Align the names and notations.
  3. [Section V-B.1, Figure 3] The bar chart text is not legible in the PDF and the numbers are not referenced; also the claim about SourceP instability is contradicted by Table III, as noted in the major comments.
  4. [Equation (15)] The pseudo-label loss uses the predicted probability yhat_unlabeled,j as both the target and the log input; this is unusual. Please clarify whether hard pseudo-labels are used, and if so, define the loss with the hard label.
  5. [References] Reference [13] is a duplicate of [12]; reference [14] is listed as an ACM URL but is not properly formatted; several in-text citations appear as '[?]' (e.g., in Section I and Section VI). The reference list also contains unrelated entries (e.g., flood susceptibility modeling).
  6. [Section V-B.4, Figure 5] The ablation results are only in a radar figure with garbled text; the exact F1/Precision/Recall values are not given. Report the numeric table for all ablation combinations so the claims can be verified.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central F1 claim is tested on held-out labels against external baselines; the few self-citations are peripheral and not load-bearing.

full rationale

I walked the derivation chain and found no step where a 'prediction' reduces by construction to a fitted input or to a self-citation. The contrastive objective (Eq. 12) is a standard InfoNCE-style loss over augmented views of unlabeled contracts; it does not incorporate test labels. The classifier is trained on labeled and pseudo-labeled data (Eqs. 13-18), and the headline results in Table III are reported on a held-out test split (contracts 251-341 plus remaining non-Ponzi contracts). The comparison baselines (Ridge-NC, SVM-NC, XGBoost-TF-IDF, MulCas, SadPonzi, SourceP) are external prior methods, not the authors' own fitted parameters. The multi-vector cosine similarity (Eq. 11 and Section IV) is a new aggregation rule for three feature vectors; its validation against centroid and weighted-average variants is an ablation, not a circular reuse of the target result. The paper contains several self-citations (e.g., refs. [81], [95]-[99]), but they appear in introductory, related-work, and future-work contexts and do not carry any load-bearing uniqueness or existence claim. I also note, as non-circularity concerns, that the manuscript has missing reference placeholders ('Wang et al. [ ?]' in Section I and 'Lu et al. [ ?]' in Section V-B), reports no overlap check between the 10,051-contract pre-training corpus and the evaluation datasets, and the medium augmentation of Eq. 6 replaces function bodies with return statements, which would remove payout logic; these are correctness, reproducibility, and potential data-leakage risks, not circular reductions. Because the central evaluation is self-contained against external benchmarks, the circularity score is low.

Assumptions & free parameters 5 free parameters · 5 assumptions · 1 invented entities

The central claim rests on several hand-set hyperparameters and on assumptions about the pre-training corpus and augmentation semantics that the paper does not verify. The only invented entity is the intermediate vector v, a mathematical construct used to aggregate three views. The proof of existence in Section IV-A is invalid as written, though a correct existence argument follows from the linear system in Section IV-B.

free parameters (5)
  • lambda1 = 1.0
    Weight for supervised loss in Lclass (Eq. 18), chosen by hand; no sensitivity analysis.
  • lambda2 = 0.85
    Weight for pseudo-label loss in Lclass (Eq. 18), chosen by hand.
  • tau = 2
    Temperature in contrastive loss (Eq. 12), chosen by hand.
  • k = 4
    Number of sub-variables in strong augmentation (Eq. 4), random integer between 2 and 5; paper reports k=4, no sensitivity.
  • theta = not specified
    Confidence threshold for pseudo-label selection (Eq. 14); listed in symbol table but no value or tuning procedure given in Section V-A.
assumptions (5)
  • standard math A homogeneous 2x3 linear system has a nonzero solution (nullity at least 1).
    Used in Section IV-B to show the intermediate vector v exists.
  • domain assumption The three data augmentations (variable splitting, function-body replacement, variable renaming) preserve semantics relevant to Ponzi identification.
    Section III-B defines the augmentations; no validation that they preserve task-relevant features.
  • domain assumption The 10,051-contract pre-training corpus is representative of downstream evaluation contracts and does not overlap with XBlock/EPSD/EBD/honeypot/phishing test sets.
    Section V-A.1 describes the crawled pre-training set; no overlap analysis is provided.
  • domain assumption GraphCodeBERT's pre-trained representations transfer to Solidity source code and DFGs.
    Section III-C uses GraphCodeBERT as feature extractor without fine-tuning or domain adaptation analysis.
  • ad hoc to paper Using an equal-angle intermediate vector to aggregate three views is a valid and useful similarity measure for contrastive learning.
    Section III-D and IV introduce this design; justification is limited to claims of faster convergence and robustness, with no theoretical or independent empirical support.
invented entities (1)
  • intermediate vector v with equal angles to the three augmented-view vectors
    purpose: Used to compute a group-level cosine similarity sim(alpha_s, alpha_w, alpha_m) = cos(alpha_s, v) for the triplet contrastive loss (Eqs. 11-12).
    The vector v is a construct introduced by the paper to aggregate three views; it is defined solely by the three input vectors and has no falsifiable handle beyond the paper's own experiments.

how reviews work

0 comments
Cite this review

Pith. "Pith review of CASPER: Contrastive Approach for Smart Ponzi Scheme Detecter with More Negative Samples." pith.science (2026). https://pith.science/paper/P5LCM6BW

@misc{pith2026250716840,
  author       = {Pith},
  title        = {Pith review of: CASPER: Contrastive Approach for Smart Ponzi Scheme Detecter with More Negative Samples},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P5LCM6BW}},
  note         = {Machine review of arXiv:2507.16840}
}
read the original abstract

The rapid evolution of digital currency trading, fueled by the integration of blockchain technology, has led to both innovation and the emergence of smart Ponzi schemes. A smart Ponzi scheme is a fraudulent investment operation in smart contract that uses funds from new investors to pay returns to earlier investors. Traditional Ponzi scheme detection methods based on deep learning typically rely on fully supervised models, which require large amounts of labeled data. However, such data is often scarce, hindering effective model training. To address this challenge, we propose a novel contrastive learning framework, CASPER (Contrastive Approach for Smart Ponzi detectER with more negative samples), designed to enhance smart Ponzi scheme detection in blockchain transactions. By leveraging contrastive learning techniques, CASPER can learn more effective representations of smart contract source code using unlabeled datasets, significantly reducing both operational costs and system complexity. We evaluate CASPER on the XBlock dataset, where it outperforms the baseline by 2.3% in F1 score when trained with 100% labeled data. More impressively, with only 25% labeled data, CASPER achieves an F1 score nearly 20% higher than the baseline under identical experimental conditions. These results highlight CASPER's potential for effective and cost-efficient detection of smart Ponzi schemes, paving the way for scalable fraud detection solutions in the future.

Figures

Figures reproduced from arXiv: 2507.16840 by the authors.

Figure 1
Figure 1. The overall framework: Step 1: Train the model’s feature extractor with a preset self-supervised representation learning framework using unlabeled data. Step 2:The classifier is trained jointly using labeled and unlabeled data by extracting features through a feature extractor. Step 3: Input the unknown smart contract into the trained feature encoder to extract the features and input them into the trained classifier… view at source ↗
Figure 2
Figure 2. Schematic diagram used to prove the existence of vectors. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. A bar chart depicting the F1 score and Acc results between CASPER [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Accuracy and training time for various similarity calculation methods [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Radar graphs of ablation experiments for each module of the model. [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Classification of Smart Ponzi Scheme Detection Methods [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

94 extracted references · 74 canonical work pages

  1. [1]

    Detecting Ponzi Schemes on Ethereum: Towards Healthier Blockchain Technology

    Weili Chen, Zibin Zheng, Jiahui Cui, Edith Ngai, Peilin Zheng, and Yuren Zhou. Detecting Ponzi Schemes on Ethereum: Towards Healthier Blockchain Technology. In Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW ’18 , pages 1409–1418. ACM Press, 2018

    2018

  2. [2]

    Khoshgoftaar

    Safak Kayikci and Taghi M. Khoshgoftaar. Blockchain meets machine learning: a survey. Journal of Big Data , 11:9, 2024

    2024

  3. [3]

    An Overview of Blockchain Technology: Architecture, Consensus, and Future Trends

    Zibin Zheng, Shaoan Xie, Hongning Dai, Xiangping Chen, and Huaimin Wang. An Overview of Blockchain Technology: Architecture, Consensus, and Future Trends. In 2017 IEEE International Congress on Big Data (BigData Congress) , pages 557–564. IEEE, 2017

    2017

  4. [4]

    AND ZHENG , Z

    C HEN , W. AND ZHENG , Z. AND CUI, J. AND NGAI , E. AND ZHENG , P. AND ZHOU , Y.. Detecting Ponzi Schemes on Ethereum: Towards Healthier Blockchain Technology. ACM Transactions on Software Engineering and Methodology (TOSEM) , 32(5):1–28, 2018. https: //doi.org/10.1145/3262777

    work page doi:10.1145/3262777 2018

  5. [8]

    The Blockchain Data Platform

    C HAINALYSIS . The Blockchain Data Platform. https://www. chainalysis.com/, 2024-08-12

    2024

  6. [9]

    SEC Charges Eleven Individuals in $300 Million Crypto Pyra- mid Scheme

    SEC. SEC Charges Eleven Individuals in $300 Million Crypto Pyra- mid Scheme. https://www.sec.gov/newsroom/press-releases/2022-134, 2022-08-12

    2022

  7. [11]

    Tree-sitter—Introduction

    T REE -SITTER . Tree-sitter—Introduction. https://tree-sitter.github.io/ tree-sitter/, 2024-08-16

    2024

  8. [12]

    tree-sitter/tree-sitter-javascript: Javascript grammar for tree-sitter

    T REE -SITTER JAVASCRIPT . tree-sitter/tree-sitter-javascript: Javascript grammar for tree-sitter. https://github.com/tree-sitter/ tree-sitter-javascript, 2024-08-16

    2024

Show all 94 references
  1. [13]

    tree-sitter/tree-sitter-javascript

    T REE -SITTER JAVASCRIPT . tree-sitter/tree-sitter-javascript. https: //github.com/tree-sitter/tree-sitter-javascript, 2024-08-16

    2024

  2. [14]

    Securing the Ethereum from Smart Ponzi Schemes: Identifica- tion Using Static Features

    ACM. Securing the Ethereum from Smart Ponzi Schemes: Identifica- tion Using Static Features. https://dl.acm.org/doi/full/10.1145/3571847, 2024-08-16

    2024 doi

  3. [16]

    Exploiting Blockchain Data to Detect Smart Ponzi Schemes on Ethereum

    IEEE. Exploiting Blockchain Data to Detect Smart Ponzi Schemes on Ethereum. https://ieeexplore.ieee.org/abstract/document/8668768, 2024-08-16

    2024

  4. [17]

    Source Code Vulnerability Detection: Combining Code Language Models and Code Property Graphs

    Ruitong Liu, Yanbin Wang, Haitao Xu, Bin Liu, Jianguo Sun, Zhenhao Guo, and Wenrui Ma. Source Code Vulnerability Detection: Combining Code Language Models and Code Property Graphs. arXiv:2404.14719, 2024

    2024 arXiv

  5. [18]

    Zhang, Mike Papadakis, Maxime Cordy, Xiaofei Xie, and Yves Le Traon

    Wei Ma, Mengjie Zhao, Ezekiel Soremekun, Qiang Hu, Jie M. Zhang, Mike Papadakis, Maxime Cordy, Xiaofei Xie, and Yves Le Traon. GraphCode2Vec: generic code embedding via lexical and program dependence analyses. In Proceedings of the 19th International Confer- ence on Mining Sof...

    2022

  6. [20]

    SourceP: Detecting Ponzi Schemes on Ethereum with Source Code

    Pengcheng Lu, Liang Cai, and Keting Yin. SourceP: Detecting Ponzi Schemes on Ethereum with Source Code. arXiv:2306.01665, 2024. JOURNAL OF LATEX CLASS FILES, VOL. 18, NO. 9, SEPTEMBER 2020 13

    2024 arXiv

  7. [21]

    Exploring Data Augmenta- tion for Code Generation Tasks

    Pinzhen Chen and Gerasimos Lampouras. Exploring Data Augmenta- tion for Code Generation Tasks. arXiv:2302.03499, 2023

    2023 arXiv

  8. [22]

    Source Code Data Augmentation for Deep Learning: A Survey

    Terry Yue Zhuo, Zhou Yang, Zhensu Sun, Yufei Wang, Li Li, Xiaoning Du, Zhenchang Xing, and David Lo. Source Code Data Augmentation for Deep Learning: A Survey. arXiv:2305.19915, 2023

    2023 arXiv

  9. [23]

    Ponzi Scheme Detection in Smart Contract via Transaction Semantic Representation Learning

    Jie Cai, Bin Li, Jiale Zhang, and Xiaobing Sun. Ponzi Scheme Detection in Smart Contract via Transaction Semantic Representation Learning. IEEE Transactions on Reliability , 73(2):1117–1131, 2024

    2024

  10. [24]

    GraphMoco: a Graph Momentum Contrast Model that Using Multimodel Structure Information for Large-scale Binary Function Representation Learning

    Sun Runjin, Guo ShiZe, Li Wei, Zhang XingYu, Guo Xi, and Pan ZhiSong. GraphMoco: a Graph Momentum Contrast Model that Using Multimodel Structure Information for Large-scale Binary Function Representation Learning. arXiv:2305.10826, 2023

    2023 arXiv

  11. [25]

    Nghi D. Q. Bui, Yijun Yu, and Lingxiao Jiang. Self-Supervised Con- trastive Learning for Code Retrieval and Summarization via Semantic- Preserving Transformations. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieva...

    2021

  12. [26]

    Self-supervised learning of smart contract representations

    Shouliang Yang, Xiaodong Gu, and Beijun Shen. Self-supervised learning of smart contract representations. In Proceedings of the 30th IEEE/ACM International Conference on Program Comprehension, pages 82–93. ACM, 2022

    2022

  13. [27]

    Clus- tering the Source Code

    Nadim Asif, Faisal Shahzad, Najia Saher, and Waseem Nazar. Clus- tering the Source Code. ResearchGate, 2009

    2009

  14. [28]

    Blockchain-Enabled Smart Contracts: Architecture, Applications, and Future Trends

    Shuai Wang, Liwei Ouyang, Yong Yuan, Xiaochun Ni, Xuan Han, and Fei-Yue Wang. Blockchain-Enabled Smart Contracts: Architecture, Applications, and Future Trends. IEEE Transactions on Systems, Man, and Cybernetics: Systems , 49(11):2266–2277, 2019

    2019

  15. [29]

    Smart Contract Development: Challenges and Opportunities

    Weiqin Zou, David Lo, Pavneet Singh Kochhar, Xuan-Bach Dinh Le, Xin Xia, Yang Feng, Zhenyu Chen, and Baowen Xu. Smart Contract Development: Challenges and Opportunities. IEEE Transactions on Software Engineering, 47(10):2084–2106, 2021

    2021

  16. [30]

    SADPonzi: Detecting and Characterizing Ponzi Schemes in Ethereum Smart Contracts

    Weimin Chen, Xinran Li, Yuting Sui, Ningyu He, Haoyu Wang, Lei Wu, and Xiapu Luo. SADPonzi: Detecting and Characterizing Ponzi Schemes in Ethereum Smart Contracts. Proceedings of the ACM on Measurement and Analysis of Computing Systems , 5(2):1–30, 2021

    2021

  17. [31]

    Blockchain challenges and opportunities: A survey

    Zibin Zheng, Shaoan Xie, Hong-Ning Dai, Xiangping Chen, and Huaimin Wang. Blockchain challenges and opportunities: A survey. International Journal of Web and Grid Services , 14(4):352–375, 2018

    2018

  18. [32]

    An overview of smart contract and use cases in blockchain technology

    Bhabendu Kumar Mohanta, Soumyashree S Panda, and Debasish Jena. An overview of smart contract and use cases in blockchain technology. In 2018 9th International Conference on Computing, Communication and Networking Technologies (ICCCNT) , pages 1–4. IEEE, 2018

    2018

  19. [33]

    Contract law 2.0: ‘Smart’ contracts as the begin- ning of the end of classic contract law

    Alexander Savelyev. Contract law 2.0: ‘Smart’ contracts as the begin- ning of the end of classic contract law. Information & Communications Technology Law, 26(2):116–134, 2017

    2017

  20. [34]

    A vademecum on blockchain technologies: When, which, and how

    Marianna Belotti, Nikola Bo ˇzi´c, Guy Pujolle, and Stefano Secci. A vademecum on blockchain technologies: When, which, and how. IEEE Communications Surveys & Tutorials , 21(4):3796–3838, 2019

    2019

  21. [35]

    Systematic review of security vulnerabilities in ethereum blockchain smart contract

    Satpal Singh Kushwaha, Sandeep Joshi, Dilbag Singh, Manjit Kaur, and Heung-No Lee. Systematic review of security vulnerabilities in ethereum blockchain smart contract. IEEE Access , 10:6605–6621, 2022

    2022

  22. [36]

    An overview on smart contracts: Challenges, advances and platforms

    Zibin Zheng, Shaoan Xie, Hong-Ning Dai, Weili Chen, Xiangping Chen, Jian Weng, and Muhammad Imran. An overview on smart contracts: Challenges, advances and platforms. Future Generation Computer Systems, 105:475–491, 2020

    2020

  23. [37]

    Decentralized applications: The blockchain- empowered software system

    Wei Cai, Zehua Wang, Jason B Ernst, Zhen Hong, Chen Feng, and Victor CM Leung. Decentralized applications: The blockchain- empowered software system. IEEE Access, 6:53019–53033, 2018

    2018

  24. [38]

    Al-SPSD: Anti-leakage smart Ponzi schemes detection in blockchain

    Shuhui Fan, Shaojing Fu, Haoran Xu, and Xiaochun Cheng. Al-SPSD: Anti-leakage smart Ponzi schemes detection in blockchain. Information Processing & Management , 58(4):102587, 2021

    2021

  25. [39]

    Securing the ethereum from smart ponzi schemes: Identification using static features

    Zibin Zheng, Weili Chen, Zhijie Zhong, Zhiguang Chen, and Yutong Lu. Securing the ethereum from smart ponzi schemes: Identification using static features. ACM Transactions on Software Engineering and Methodology, 32(5):1–28, 2023

    2023

  26. [40]

    Criminological Theory: Context and Consequences

    J Robert Lilly, Francis T Cullen, and Richard A Ball. Criminological Theory: Context and Consequences . Sage Publications, 2018

    2018

  27. [41]

    Ponzi schemes: A critical analysis

    Surendranath Rakesh Jory and Mark J Perry. Ponzi schemes: A critical analysis. SSRN:1894206, 2011

    2011

  28. [42]

    Analyzing the Bitcoin Ponzi scheme ecosystem

    Marie Vasek and Tyler Moore. Analyzing the Bitcoin Ponzi scheme ecosystem. In Financial Cryptography and Data Security , pages 101–

  29. [43]

    Ponzi schemes and its prevention: Insights from Malaysia

    Eley Suzana Kasim, Norlaila Md Zina, Hazlina Mohd Padil, and Normah Omar. Ponzi schemes and its prevention: Insights from Malaysia. Management & Accounting Review , 19(3):89–118, 2020

    2020

  30. [44]

    A feature- based robust method for abnormal contracts detection in ethereum blockchain

    Ali Aljofey, Abdur Rasool, Qingshan Jiang, and Qiang Qu. A feature- based robust method for abnormal contracts detection in ethereum blockchain. Electronics, 11(18):2937, 2022

    2022

  31. [45]

    BAI Saga: Pyramid Scheme, Ponzi Scheme, Ponzi-like Scheme or Political Vendetta and Conspiracy? 2023

    Guru Dev Teeluckdharry. BAI Saga: Pyramid Scheme, Ponzi Scheme, Ponzi-like Scheme or Political Vendetta and Conspiracy? 2023

    2023

  32. [46]

    Dissecting Ponzi schemes on Ethereum: identification, analysis, and impact

    Massimo Bartoletti, Salvatore Carta, Tiziana Cimoli, and Roberto Saia. Dissecting Ponzi schemes on Ethereum: identification, analysis, and impact. Future Generation Computer Systems , 102:259–277, 2020

    2020

  33. [47]

    Modeling flood susceptibility using data-driven approaches of na ¨ıve bayes tree, alternating decision tree, and random forest methods

    Wei Chen, Yang Li, Weifeng Xue, Himan Shahabi, Shaojun Li, Haoyuan Hong, Xiaojing Wang, Huiyuan Bian, Shuai Zhang, Biswa- jeet Pradhan, et al. Modeling flood susceptibility using data-driven approaches of na ¨ıve bayes tree, alternating decision tree, and random forest methods...

    2020

  34. [48]

    Bitcoin theft detection based on supervised machine learning algorithms

    Binjie Chen, Fushan Wei, and Chunxiang Gu. Bitcoin theft detection based on supervised machine learning algorithms. Security and Communication Networks, 2021:6643763, 2021

    2021

  35. [49]

    Improving Ponzi scheme contract detection using multi-channel TextCNN and transformer

    Yizhou Chen, Heng Dai, Xiao Yu, Wenhua Hu, Zhiwen Xie, and Cheng Tan. Improving Ponzi scheme contract detection using multi-channel TextCNN and transformer. Sensors, 21(19):6417, 2021

    2021

  36. [50]

    Ponzi scheme detection via oversampling-based Long Short-Term Memory for smart contracts

    Lei Wang, Hao Cheng, Zibin Zheng, Aijun Yang, and Xiaohu Zhu. Ponzi scheme detection via oversampling-based Long Short-Term Memory for smart contracts. Knowledge-Based Systems, 228:107312, 2021

    2021

  37. [51]

    A simple framework for contrastive learning of visual rep- resentations

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. A simple framework for contrastive learning of visual rep- resentations. In International Conference on Machine Learning , pages 1597–1607. PMLR, 2020

    2020

  38. [52]

    Unsupervised learning of visual features by contrasting cluster assignments

    Mathilde Caron, Ishan Misra, Julien Mairal, Priya Goyal, Piotr Bo- janowski, and Armand Joulin. Unsupervised learning of visual features by contrasting cluster assignments. In Advances in Neural Information Processing Systems, volume 33, pages 9912–9924, 2020

    2020

  39. [53]

    TimesURL: Self-supervised contrastive learning for universal time series representation learning

    Jiexi Liu and Songcan Chen. TimesURL: Self-supervised contrastive learning for universal time series representation learning. In Proceed- ings of the AAAI Conference on Artificial Intelligence , volume 38, pages 13918–13926, 2024

    2024

  40. [54]

    MixCon3D: Synergizing Multi-View and Cross-Modal Contrastive Learning for Enhancing 3D Representation

    Yipeng Gao, Zeyu Wang, Wei-Shi Zheng, Cihang Xie, and Yuyin Zhou. MixCon3D: Synergizing Multi-View and Cross-Modal Contrastive Learning for Enhancing 3D Representation. arXiv:2311.01734, 2023

    2023 arXiv

  41. [55]

    Multi- scale subgraph contrastive learning

    Yanbei Liu, Yu Zhao, Xiao Wang, Lei Geng, and Zhitao Xiao. Multi- scale subgraph contrastive learning. arXiv:2403.02719, 2024

    2024 arXiv

  42. [56]

    tree-sitter-solidity—Introduction

    T REE -SITTER SOLIDITY . tree-sitter-solidity—Introduction. https: //github.com/JoranHonig/tree-sitter-solidity, 2024-09-01

    2024

  43. [57]

    Graphcodebert: Pre-training code representations with data flow

    Daya Guo, Shuo Ren, Shuai Lu, Zhangyin Feng, Duyu Tang, Shujie Liu, Long Zhou, Nan Duan, Alexey Svyatkovskiy, Shengyu Fu, et al. Graphcodebert: Pre-training code representations with data flow. arXiv:2009.08366, 2020

    2009 arXiv

  44. [58]

    https://etherscan.io/, 2024-09-01

    E THERSCAN . https://etherscan.io/, 2024-09-01

    2024

  45. [59]

    https://www.blockscout.com/, 2024-09-01

    B LOCKSCOUT . https://www.blockscout.com/, 2024-09-01

    2024

  46. [60]

    https://bscscan.com/, 2024-09-01

    B SCSCAN . https://bscscan.com/, 2024-09-01

    2024

  47. [61]

    Explainable Ponzi Schemes Detection on Ethereum

    Letterio Galletta and Fabio Pinelli. Explainable Ponzi Schemes Detection on Ethereum. In Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing , pages 1014–1023, 2024

    2024

  48. [62]

    CS4545/CS6545 Project Report: Clustering Solidity Smart Contracts by Similarity

    Ansumana F Jadama and Aditya Dilip Thakur. CS4545/CS6545 Project Report: Clustering Solidity Smart Contracts by Similarity. https://www. researchgate.net/publication/381773424

  49. [63]

    https://bitcointalk.org/

    B ITCOIN FORUM . https://bitcointalk.org/

  50. [64]

    Nakamoto

    S. Nakamoto. Bitcoin: A Peer-to-Peer Electronic Cash System. 2008. https://bitcoin.org/bitcoin.pdf

    2008

  51. [65]

    M. Swan. Blockchain: Blueprint for a New Economy . O’Reilly Media, 2015

    2015

  52. [66]

    G. Wood. Ethereum: A Secure Decentralised Generalised Transaction Ledger. Ethereum Project Yellow Paper, 2014. https://ethereum.github. io/yellowpaper/paper.pdf

    2014

  53. [67]

    V . Buterin. Next Generation Smart Contract & Decentralized Appli- cation Platform. White Paper, 2014. https://github.com/ethereum/wiki/ wiki/White-Paper

    2014

  54. [68]

    W. Cai, Z. Wang, X. Li, A. Vinel, X. Zhang, and Y . Fang. Decentralized Applications: The Blockchain-Empowered Software Systems. IEEE Access, 6:53019–53033, 2018

    2018

  55. [69]

    Meadow, N

    D. Meadow, N. Shadbolt, and L. Burkhalter. Ponzi Schemes and the Blockchain. arXiv:1802.03628, 2018

    2018 arXiv

  56. [70]

    Bartoletti and L

    M. Bartoletti and L. Pompianu. An Empirical Analysis of Ponzi Schemes in the Bitcoin Domain. arXiv:1704.00756, 2017

    2017 arXiv

  57. [71]

    2019 Crypto Crime Report

    C HAINALYSIS . 2019 Crypto Crime Report. https://www.chainalysis. com/reports/cryptocrimereport, 2019. JOURNAL OF LATEX CLASS FILES, VOL. 18, NO. 9, SEPTEMBER 2020 14

    2019

  58. [72]

    D. He, K. Wang, and L. Guo. EoSafe: Detecting Vulnerabilities in EOSIO Smart Contracts. IEEE Transactions on Dependable and Secure Computing, 18(4):1519–1532, 2021

    2021

  59. [73]

    S. Tang, Y . Wang, and J. Li. A Survey on Blockchain Technology and its Application in IoT. IEEE Internet of Things Journal , 11(4):2778– 2792, 2024

    2024

  60. [74]

    Fazel, M

    N. Fazel, M. Khan, and M. Conti. Security and Privacy in IoT: A Survey. ACM Transactions on Internet of Things , 3(2):1–26, 2024

    2024

  61. [75]

    A. Afaq, N. Javaid, and A. Ahmad. Blockchain Technology: A Survey on Applications, Challenges, and Future Directions. IEEE Access , 12:45678–45695, 2024

    2024

  62. [76]

    Mason and H

    J. Mason and H. Escott. Smart contracts in construction: Views and perceptions of stakeholders. Proc FIG Conf , pages 2006–2009, 2018

    2006

  63. [77]

    IDPonzi: An interpretable detection model for identifying smart Ponzi schemes

    Xia Feng, Qichen Shi, Xingye Li, Haiyang Liu, and Liangmin Wang. IDPonzi: An interpretable detection model for identifying smart Ponzi schemes. Engineering Applications of Artificial Intelligence , 136:108868, 2024

    2024

  64. [78]

    Ponzi scheme detection based on control flow graph feature extraction

    Shunhui Ji, Congxiong Huang, Pengcheng Zhang, Hai Dong, and Yan Xiao. Ponzi scheme detection based on control flow graph feature extraction. In 2023 IEEE International Conference on Web Services (ICWS), pages 585–594. IEEE, 2023

    2023

  65. [79]

    A first look at blockchain-based decentralized applications

    Kaidong Wu, Yun Ma, Gang Huang, and Xuanzhe Liu. A first look at blockchain-based decentralized applications. Software: Practice and Experience, 51(10):2033–2050, 2021

    2021

  66. [80]

    Investigating MMM Ponzi scheme on bitcoin

    Yazan Boshmaf, Charitha Elvitigala, Husam Al Jawaheri, Primal Wije- sekera, and Mashael Al Sabah. Investigating MMM Ponzi scheme on bitcoin. In Proceedings of the 15th ACM Asia Conference on Computer and Communications Security , pages 519–530, 2020

    2020

  67. [81]

    Transaction-based classification and detection approach for Ethereum smart contract

    Teng Hu, Xiaolei Liu, Ting Chen, Xiaosong Zhang, Xiaoming Huang, Weina Niu, Jiazhong Lu, Kun Zhou, and Yuan Liu. Transaction-based classification and detection approach for Ethereum smart contract. Information Processing & Management , 58(2):102462, 2021

    2021

  68. [82]

    Detection of Ponzi scheme on Ethereum using machine learning algorithms

    Ifeyinwa Jacinta Onu, Abiodun Esther Omolara, Moatsum Alawida, Oludare Isaac Abiodun, and Abdulatif Alabdultif. Detection of Ponzi scheme on Ethereum using machine learning algorithms. Scientific Reports, 13(1):18403, 2023

    2023

  69. [83]

    Evaluating machine- learning techniques for detecting smart ponzi schemes

    Giacomo Ibba and Giuseppe Antonio Pierro. Evaluating machine- learning techniques for detecting smart ponzi schemes. In 2021 IEEE/ACM 4th International Workshop on Emerging Trends in Soft- ware Engineering for Blockchain (WETSEB), pages 34–40. IEEE, 2021

    2021

  70. [84]

    Data-driven smart ponzi scheme detection

    Yuzhi Liang, Weijing Wu, Kai Lei, and Feiyang Wang. Data-driven smart ponzi scheme detection. arXiv:2108.09305, 2021

    2021 arXiv

  71. [85]

    Ponzi Scheme Detection Based on CNN and BiGRU combined with Attention Mechanism

    Bo Cui and Guoqing Wang. Ponzi Scheme Detection Based on CNN and BiGRU combined with Attention Mechanism. In 2024 27th International Conference on Computer Supported Cooperative Work in Design (CSCWD) , pages 1852–1857. IEEE, 2024

    2024

  72. [86]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervision. In International Conference on Machine Learning, pa...

    2021

  73. [87]

    ULIP: Learning a unified representation of language, images, and point clouds for 3D understanding

    Le Xue, Mingfei Gao, Chen Xing, Roberto Mart ´ın-Mart´ın, Jiajun Wu, Caiming Xiong, Ran Xu, Juan Carlos Niebles, and Silvio Savarese. ULIP: Learning a unified representation of language, images, and point clouds for 3D understanding. In Proceedings of the IEEE/CVF Conference o...

    2023

  74. [88]

    A correlated topic model of science

    David M Blei and John D Lafferty. A correlated topic model of science. 2007

    2007

  75. [89]

    A hierarchical clustering algorithm using dynamic modeling

    George Karypis, Euihong Han, and Vipin Kumar. A hierarchical clustering algorithm using dynamic modeling. 1999

    1999

  76. [90]

    SoRec: social recommendation using probabilistic matrix factorization

    Hao Ma, Haixuan Yang, Michael R Lyu, and Irwin King. SoRec: social recommendation using probabilistic matrix factorization. In Pro- ceedings of the 17th ACM Conference on Information and Knowledge Management, pages 931–940, 2008

    2008

  77. [91]

    The art of the scam: Demystifying honeypots in ethereum smart contracts

    Christof Ferreira Torres and Mathis Steichen. The art of the scam: Demystifying honeypots in ethereum smart contracts. In 28th USENIX Security Symposium, pages 1591–1607, 2019

    2019

  78. [92]

    April 2024 Ponzi Scheme Roundup

    The Ponzi Scheme Blog. April 2024 Ponzi Scheme Roundup. https://theponzibook.blogspot.com/2024/04/ april-2024-ponzi-scheme-roundup.html, 2025-01-03

    2024

  79. [93]

    XGBoost: A scalable tree boosting system

    Tianqi Chen and Carlos Guestrin. XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , pages 785– 794, 2016

    2016

  80. [94]

    Multilayer perceptron (MLP)

    Hind Taud and Jean-Franc ¸ois Mas. Multilayer perceptron (MLP). In Geomatic Approaches for Modeling Land Change Scenarios , pages 451–455. Springer, 2017

    2017

  81. [95]

    Fuzzing: A Survey for Roadmap

    Xiaogang Zhu, Sheng Wen, Seyit Camtepe, and Yang Xiang. Fuzzing: A Survey for Roadmap. ACM Computing Surveys, 54(11s):230, 2022

    2022

  82. [96]

    When Software Security Meets Large Language Mod- els: A Survey

    Xiaogang Zhu, Wei Zhou, Qing-Long Han, Wanlun Ma, Sheng Wen, and Yang Xiang. When Software Security Meets Large Language Mod- els: A Survey. IEEE/CAA Journal of Automatica Sinica , 12(2):317– 334, 2025

    2025

  83. [97]

    The Security of Using Large Language Models—A Survey with Emphasis on ChatGPT

    Wei Zhou, Xiaogang Zhu, Qing-Long Han, Lin Li, Xiao Chen, Sheng Wen, and Yang Xiang. The Security of Using Large Language Models—A Survey with Emphasis on ChatGPT. IEEE/CAA Journal of Automatica Sinica , 12(1):1–26, 2025

    2025

  84. [98]

    A Blockchain-Based Decentralized, Fair and Authenticated Information Sharing Scheme in Zero Trust Internet-of-Things

    Yizhi Liu, Xiaohan Hao, Wei Ren, Ruoting Xiong, Tianqing Zhu, Kim-Kwang Raymond Choo, and Geyong Min. A Blockchain-Based Decentralized, Fair and Authenticated Information Sharing Scheme in Zero Trust Internet-of-Things. IEEE Transactions on Computers , 72(2):501–512, 2023

    2023

  85. [99]

    Federated Learning with Blockchain-Enhanced Machine Unlearning: A Trustworthy Approach

    Xuhan Zuo, Minghao Wang, Tianqing Zhu, Lefeng Zhang, Shui Yu, and Wanlei Zhou. Federated Learning with Blockchain-Enhanced Machine Unlearning: A Trustworthy Approach. IEEE Transactions on Services Computing, pages 1–15, 2025. Weijia Yang graduated from Chengdu University of In...

    2025

  86. [2004]

    in Information and Communication Engineering

    From March 2004 to December 2010, he stud- ied for a Ph.D. in Information and Communication Engineering. JOURNAL OF LATEX CLASS FILES, VOL. 18, NO. 9, SEPTEMBER 2020 15 Tianqing Zhu is a professor at City University of Macau, before that, she was a lecturer at Deakin Universit...

    2004

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.