Pith. sign in

REVIEW 3 major objections 4 minor 85 references

Blockchain Data Analysis in the Era of Large-Language Models

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This position paper argues that large language models, with pre-trained knowledge, cross-chain generalizability, and explainable outputs, can overcome the three core limits of blockchain data analysis: data scarcity, protocol…

desk verdict A useful but shaky position paper: the four-pattern framework is a reasonable organizing device, but the core claim that pre-trained LLMs solve blockchain data scarcity is undercut by the paper's own cited examples, and the citation errors need fixing. read the letter →

arxiv 2412.09640 v1 pith:I6VHYQNO submitted 2024-12-09 cs.CR cs.AI

Kentaroh Toyoda , Xiao Wang , Mingzhe Li , Bo Gao , Yuan Wang , Qingsong Wei This is my paper · ORCID
classification cs.CRcs.AI
keywords blockchaindataanalysislargelanguagemodelspromptengineeringretrieval-augmentedgenerationdesignpatternsfrauddetectionsmartcontractauditingcross-chaingeneralizability
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 argues that large language models (LLMs) can become a core technology for blockchain data analysis by addressing three long-standing bottlenecks: the scarcity of labeled ground truth, the need to re-engineer tools for each blockchain protocol, and the difficulty of explaining analytical results to auditors, regulators, and developers. To make this concrete, it organizes the landscape into data sources, downstream tasks such as fraud detection, smart contract auditing, market prediction, governance and compliance monitoring, and privacy analysis, and it proposes four reusable design patterns for wiring LLMs into analytic pipelines. The intended payoff is a systematic framework that lets practitioners choose where the LLM should sit, whether as a feature extractor feeding a traditional classifier, as an interpreter of a traditional model's output, as an end-to-end predictor, or as a multi-stage reasoning chain. The paper also flags latency, reliability, cost, scalability, generalizability, and autonomy as the open problems that will decide whether this vision holds.

What carries the argument

The load-bearing machinery is the pair of prompt-engineering techniques and the four design patterns. Prompt engineering supplies the controls: instructions and context that frame the task, in-context learning that injects demonstrations, retrieval-augmented generation that pulls relevant external knowledge into the prompt, reasoning frameworks such as Chain-of-Thought, Tree-of-Thought, and Graph-of-Thought that force stepwise or branching reasoning, and compression strategies that fit large transaction histories into token limits. The design patterns specify where the LLM sits in the pipeline: as a preprocessor (Pattern 1), as an interpreter of a traditional model's output (Pattern 2), as a direct end-to-end predictor (Pattern 3), or as part of a multi-stage LLM chain (Pattern 4). Together they convert the abstract claim that LLMs help into a decision procedure: choose a pattern based on whether the task needs feature enrichment, explanation, direct prediction, or iterative reasoning, then apply the appropriate prompting technique to control the input and output.

What would settle it

A concrete test: take a general-purpose LLM and, using only zero-shot prompting with the paper's template, classify a held-out set of labeled Ethereum phishing and Ponzi addresses from a public benchmark; compare its F1 against a trained graph neural network baseline. If the zero-shot LLM performs at or near chance or produces confident but incorrect rationales, the core transfer claim and Pattern 3's viability are falsified for the most safety-critical task. A second test would fine-tune on Ethereum and evaluate on Solana to see whether cross-chain generalizability actually holds without re-engineering.

Watch

Extended reading notes

Core claim

The paper's central claim is that LLMs' pre-trained knowledge, generalizability across heterogeneous chains, and explainability can mitigate the three main weaknesses of current blockchain analytics: pseudonymity-coupled fraud, absence of labeled datasets, protocol fragmentation, and opaque insights (Section 2.4). It asserts that prompt engineering techniques, in-context learning, retrieval-augmented generation, reasoning frameworks, and input compression, combined with the four design patterns, constitute a systematic way to integrate LLMs into blockchain analytics. The four patterns are presented as an architectural vocabulary: Pattern 1 treats the LLM as an enhancer that transforms raw data into enriched features for a traditional predictor; Pattern 2 applies a traditional predictor first and lets the LLM refine or explain its output; Pattern 3 uses the LLM directly as the final predictor; and Pattern 4 chains two LLM stages through an intermediate representation. The paper's position is that this vocabulary, together with the use cases it surveys, shows LLM integration is not a collection of one-off demos but a coherent design space worth systematic exploitation.

Load-bearing premise

The argument rests on the premise that an LLM's pre-trained knowledge and reasoning transfer to blockchain data without extensive per-chain fine-tuning, so that it can interpret transaction semantics, explain its outputs, and generalize across protocols; if that transfer fails, every pattern inherits the failure, and Pattern 3's direct prediction is the most exposed.

Editorial extensions

If this is right

  • Fraud detection and compliance monitoring could produce risk scores accompanied by human-readable rationales, even when labeled training data for new scam types does not yet exist.
  • Analytic tools built on the enhancer patterns could span multiple chains, such as Ethereum and Solana, without per-protocol feature engineering, because the LLM absorbs the protocol differences in context.
  • Smart-contract auditing could combine static analyzers with an LLM verifier layer, improving precision and recall over either tool alone.
  • The multi-stage pattern enables complex workflows where one LLM builds an intermediate representation, such as a call graph or annotated control-flow graph, and a second LLM reasons over it for vulnerability detection or trading decisions.
  • If the autonomy challenge is solved, AI agents could assemble these workflows on demand, automating monitoring, fraud detection, and trading with minimal human intervention.

Reading between the lines

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

  • A direct test of the paper's core bet would be a standardized zero-shot benchmark: run a general-purpose LLM, without fine-tuning, on labeled Ethereum phishing and Ponzi datasets and compare its F1 against trained graph neural networks; near-chance performance on Pattern 3 would call the transfer premise into question.
  • The RAG suggestion points to an obvious extension the paper leaves implicit: grounding LLM outputs in a retrievable knowledge base of verified blockchain threat intelligence, standards, and past audit findings could be the most direct route to reducing hallucination in compliance settings.
  • Pattern 2 may be the more defensible default for high-stakes regulatory use, since a deterministic traditional model provides the numeric output and the LLM only explains it, limiting the blast radius of an LLM error.
  • Benchmarking the same prompt on UTXO-based and account-based chains would isolate whether the claimed generalizability comes from pre-trained knowledge or from the prompt template, informing where the framework's boundaries lie.
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

3 major / 4 minor

Summary. This paper is a position paper and survey on using large language models for blockchain data analysis. It catalogues on-chain and off-chain data types, downstream tasks (fraud detection, smart contract analysis, market prediction, governance/compliance monitoring, and privacy analysis), and conventional analytical techniques. It identifies data scarcity, protocol fragmentation, scalability, and interpretability as the main challenges, then argues that LLMs' pre-trained knowledge, cross-chain generalizability, and explainability can mitigate these challenges. The paper proposes prompt-engineering elements and four design patterns adapted from a time-series LLM taxonomy: LLM as feature enhancer, LLM as output interpreter, LLM as direct predictor, and multi-stage LLM pipelines. It maps these patterns to use cases and closes with six future research directions: latency, reliability, cost, scalability, generalizability, and autonomy.

Significance. If the central assertion were established, this framework would give practitioners a useful vocabulary for deploying LLMs in blockchain analytics and would focus research on cost, latency, and reliability. The paper is timely, broad, and written accessibly; the prompt templates in Section 4.1 and the explicit statement that no LLM-based privacy analysis currently exists are useful contributions. However, the paper reports no experiments, benchmarks, or formal derivations, so its value is that of an opinionated taxonomy rather than an empirical demonstration. More importantly, the cited evidence for the strongest version of the claim is mixed: at least two Pattern 3 examples require domain-specific training or are described inconsistently within the paper, and several references are attributed incorrectly. These problems need to be resolved before the framework can be relied upon as a systematic survey.

major comments (3)
  1. [Section 3 and §4.2 (Pattern 3)] The central premise that LLMs' pre-trained knowledge will overcome data scarcity and enable cross-chain generalizability is not supported by the paper's own Pattern 3 examples. In §4.2, BERT4ETH [66] is presented as 'a pre-trained transformer' that directly analyzes transaction sequences, but §4.3.1 correctly states that BERT4ETH was pre-trained on Ethereum transaction data with masked address prediction; this is domain-specific pre-training, not the web-scale general knowledge invoked in Section 3. Likewise, BlockGPT is described in §4.2 as 'directly processing blockchain transaction traces through a pre-trained LLM,' while §4.3.4 says it 'trains an LLM from scratch' on blockchain tracing representations. These two descriptions are contradictory, and a from-scratch model cannot illustrate the claimed benefit of pre-trained knowledge under data scarcity. The paper should either replace these examples with zero-shot or few-shot applications of genuinely general-purpose LLMs to raw blockchain data, or explicitly weaken the Section 3 claim to something like 'LLMs may reduce, but not eliminate, the need for domain-specific data and training.'
  2. [§4.1 and §4.3] The paper asserts that prompt engineering techniques such as in-context learning, RAG, CoT, ToT, and GoT will improve blockchain analysis, but the provided prompts are only templates; no evidence is cited that these prompting strategies improve accuracy, calibration, or explainability on blockchain tasks. Because the main contribution is a framework for using these techniques, at least a qualitative summary of existing empirical results, or a small set of illustrative benchmarks, is needed to justify the 'enormous potential' claim in Section 1. A concrete test would be to evaluate Pattern 3 with a general-purpose LLM on standard Ethereum phishing or Ponzi datasets in zero-shot and few-shot settings, and to report whether the model's explanations are faithful rather than post-hoc.
  3. [§4.3.2 and References [67], [72]] Several key citations are incorrect, which matters for a systematic survey. The sentence 'Liu et al. introduced FELLMVP' refers to reference [67], which is by Luo et al.; the sentence 'Sun et al. developed two innovative tools: ACFIX and GPTScan [72]' cites reference [72], which is Wei et al.'s 'LLM-SmartAudit,' not the ACFIX/GPTScan papers. These errors compromise the survey's reliability as a reference work. All citations in Section 4.3 should be re-verified against the primary sources, and the reference list should be updated accordingly.
minor comments (4)
  1. [Section 3] The sentence ending 'is currentlywmissing' contains a typo; it should read 'is currently missing.'
  2. [§2.2.3] The sentence ending '(e.g., [22], [23].' is missing a closing parenthesis; it should be '(e.g., [22], [23]).'
  3. [§4.2 (Pattern 2)] The sentence 'LLM-as-predictors are systems where LLMs directly generate the final prediction or decision...' is a definition of LLM-as-predictors and appears under the 'Examples:' subheading for Pattern 2, which is confusing; it should be moved to the classification introduction or to the Pattern 3 discussion.
  4. [§4.1.4] The discussion of Tree-of-Thought and Graph-of-Thought does not cite the original works introducing those reasoning frameworks; please add the appropriate references.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is a position/survey whose claims are not derived from fitted parameters or self-citations; the BlockGPT/BERT4ETH inconsistencies are evidentiary, not circular.

full rationale

The paper is a position/survey paper, not a derivation or a fitted-model paper. Its central claim is that LLMs can mitigate data scarcity, generalizability, and explainability limitations in blockchain data analysis, and it organizes existing and proposed integrations into four design patterns. There are no equations connecting inputs to outputs, no fitted parameters, and no quantity defined in terms of another quantity, so none of the standard circularity patterns (self-definitional, fitted-input-called-prediction, uniqueness imported from authors, ansatz smuggled via citation, renaming known result) applies. The self-citations ([10], [39], [57], [76]) appear in background examples, a footnote directing readers to other surveys, and one of several examples in the reliability discussion; none of them carries the paper's central argument, so they are not load-bearing. The internal inconsistency about BlockGPT (called 'pre-trained' in Section 4.2 and described as training an LLM from scratch in Section 4.3.4) and the observation that BERT4ETH uses domain-specific pretraining rather than general-purpose LLM knowledge are legitimate evidence-quality concerns, but they do not amount to circularity: the paper's conclusions are not defined in terms of those examples. The honest finding is no significant circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim rests on assumptions about LLM capabilities and the transferability of a time-series taxonomy to blockchain data. No quantitative parameters are fitted or introduced.

assumptions (4)
  • domain assumption LLMs' pre-trained knowledge compensates for the scarcity of blockchain-specific labeled datasets.
    Invoked in Section 3.1 and throughout the use cases; no empirical support is supplied in this paper, and the paper's own cited examples involve fine-tuning or task-specific training.
  • domain assumption LLMs generalize across different blockchain protocols without extensive re-engineering.
    Stated in Section 3.2; this is the core enabler of the cross-chain claims, but it is not tested in the paper.
  • domain assumption LLM explanations are a core strength that builds trust in insights.
    Stated in Section 3.3; the paper does not address the known unreliability of LLM explanations, such as post-hoc rationalization.
  • ad hoc to paper The four design patterns from the time-series LLM taxonomy transfer to blockchain data.
    Adopted from reference [52] in Section 4.2 without validation on blockchain workloads.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Blockchain Data Analysis in the Era of Large-Language Models." pith.science (2026). https://pith.science/paper/I6VHYQNO

@misc{pith2026241209640,
  author       = {Pith},
  title        = {Pith review of: Blockchain Data Analysis in the Era of Large-Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/I6VHYQNO}},
  note         = {Machine review of arXiv:2412.09640}
}
read the original abstract

Blockchain data analysis is essential for deriving insights, tracking transactions, identifying patterns, and ensuring the integrity and security of decentralized networks. It plays a key role in various areas, such as fraud detection, regulatory compliance, smart contract auditing, and decentralized finance (DeFi) risk management. However, existing blockchain data analysis tools face challenges, including data scarcity, the lack of generalizability, and the lack of reasoning capability. We believe large language models (LLMs) can mitigate these challenges; however, we have not seen papers discussing LLM integration in blockchain data analysis in a comprehensive and systematic way. This paper systematically explores potential techniques and design patterns in LLM-integrated blockchain data analysis. We also outline prospective research opportunities and challenges, emphasizing the need for further exploration in this promising field. This paper aims to benefit a diverse audience spanning academia, industry, and policy-making, offering valuable insights into the integration of LLMs in blockchain data analysis.

Figures

Figures reproduced from arXiv: 2412.09640 by the authors.

Figure 1
Figure 1. Paper Outline. 2.2.2 Smart Contract Analysis One of the most critical tasks in smart contract analysis is the detection of vulnerabilities that could compromise security and lead to significant financial or reputational losses. Depending on the use cases, the data to be analyzed are raw source code or compiled one (bytecode) and ABI. Techniques in NLP, graph mining, and software engineering are often used for smart … view at source ↗
Figure 2
Figure 2. Illustration of How LLMs Could Solve Challenges [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. The Proposed Design Patterns that Incorporate LLMs into Blockchain Analysis. [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: An AI Agent that Automates Decision Making on [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

85 extracted references · 73 canonical work pages

  1. [67]

    FELLMVP: An Ensemble LLM Framework for Classifying Smart Contract Vulnerabilities,

    Y. Luo, W. Xu, K. Andersson, M. S. Hossain, and D. Xu, “FELLMVP: An Ensemble LLM Framework for Classifying Smart Contract Vulnerabilities,” in Proc. of IEEE International Conference on Blockchain (Blockchain), pp. 89–96, 2024

    work page 2024

  2. [72]

    LLM-SmartAudit: Advanced Smart Contract Vulnerability Detection,

    Z. Wei, J. Sun, Z. Zhang, and X. Zhang, “LLM-SmartAudit: Advanced Smart Contract Vulnerability Detection,” Oct. 2024. arXiv:2410.09381 [cs]

    arXiv 2024

  3. [66]

    BERT4ETH: A Pre-trained Transformer for Ethereum Fraud Detection,

    S. Hu, Z. Zhang, B. Luo, S. Lu, B. He, and L. Liu, “BERT4ETH: A Pre-trained Transformer for Ethereum Fraud Detection,” in Proc. of ACM Web Conference (WWW), pp. 2189–2197, Apr. 2023

    work page 2023

  4. [1]

    Traveling the Silk Road: A measurement analysis of a large anonymous online marketplace,

    N. Christin, “Traveling the Silk Road: A measurement analysis of a large anonymous online marketplace,” in Proc. of ACM Web Conference (WWW), pp. 213–224, 2013

    2013

  5. [2]

    A survey on blockchain data analy- sis,

    W. Hou, B. Cui, and R. Li, “A survey on blockchain data analy- sis,” in Proc. of IEEE Annual Computers, Software, and Applications Conference (COMPSAC), pp. 357–365, IEEE, July 2021

    2021

  6. [3]

    Blockchain data analysis from the perspective of complex networks: Overview,

    W. Song, W. Zhang, J. Wang, L. Zhai, P . Jiang, S. Huang, and B. Li, “Blockchain data analysis from the perspective of complex networks: Overview,” Tsinghua Science and Technology , vol. 28, no. 1, pp. 176–206, 2022

    2022

  7. [4]

    Visualiza- tion of blockchain data: A systematic review,

    N. Tovanich, N. Heulot, J.-D. Fekete, and P . Isenberg, “Visualiza- tion of blockchain data: A systematic review,” IEEE Transactions on Visualization and Computer Graphics, vol. 27, pp. 3135–3152, July 2021

    2021

  8. [5]

    Blockchain data mining with graph learning: A survey,

    Y. Qi, J. Wu, H. Xu, and M. Guizani, “Blockchain data mining with graph learning: A survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 46, pp. 729–748, Feb. 2024

    2024

Show all 85 references
  1. [6]

    Blockchain and machine learning for fraud detection: A privacy-preserving and adaptive incentive based approach,

    T. H. Pranto, K. T. A. M. Hasib, T. Rahman, A. B. Haque, A. K. M. N. Islam, and R. M. Rahman, “Blockchain and machine learning for fraud detection: A privacy-preserving and adaptive incentive based approach,” IEEE Access, vol. 10, pp. 87115–87134, 2022

    2022

  2. [7]

    Phishing scam detection on ethereum: Towards financial security for blockchain ecosystem,

    W. Chen, X. Guo, Z. Chen, Z. Zheng, and Y. Lu, “Phishing scam detection on ethereum: Towards financial security for blockchain ecosystem,” Proc. of International Joint Conference on Artificial Intel- ligence (IJCAI), 2020. 12

    2020

  3. [8]

    Phishing scams detection in ethereum transaction network,

    L. Chen, J. Peng, Y. Liu, J. Li, F. Xie, and Z. Zheng, “Phishing scams detection in ethereum transaction network,” ACM Transactions on Internet Technology, vol. 21, pp. 1–16, Feb. 2021

    2021

  4. [9]

    Data mining for detecting bitcoin ponzi schemes,

    M. Bartoletti, B. Pes, and S. Serusi, “Data mining for detecting bitcoin ponzi schemes,” in Proc. of Crypto Valley Conference on Blockchain Technology (CVCBT), pp. 75–84, IEEE, 2018

    2018

  5. [10]

    A novel methodology for HYIP operators’ bitcoin addresses identifica- tion,

    K. Toyoda, P . Takis Mathiopoulos, and T. Ohtsuki, “A novel methodology for HYIP operators’ bitcoin addresses identifica- tion,” IEEE Access, vol. 7, pp. 74835–74848, 2019

    2019

  6. [11]

    Machine learning methods to detect money laundering in the bitcoin blockchain in the presence of label scarcity,

    J. Lorenz, M. I. Silva, D. Apar ´ıcio, J. T. Ascens ˜ao, and P . Bizarro, “Machine learning methods to detect money laundering in the bitcoin blockchain in the presence of label scarcity,” in Proc. of ACM International Conference on AI in Finance , pp. 1–8, 2020

    2020

  7. [12]

    Graphalm: Active learning for de- tecting money laundering transactions on blockchain networks,

    Q. Wang, W.-T. Tsai, and T. Shi, “Graphalm: Active learning for de- tecting money laundering transactions on blockchain networks,” IEEE Network, 2024

    2024

  8. [13]

    Smart contract vulnerability detection using graph neural networks,

    Y. Zhuang, Z. Liu, P . Qian, Q. Liu, X. Wang, and Q. He, “Smart contract vulnerability detection using graph neural networks,” in Proc. of International Joint Conference on Artificial Intelligence (IJCAI), pp. 3283–3290, 2021

    2021

  9. [14]

    Smart con- tract vulnerability detection technique: A survey,

    P . Qian, Z. Liu, Q. He, B. Huang, D. Tian, and X. Wang, “Smart con- tract vulnerability detection technique: A survey,” arXiv preprint arXiv:2209.05872, 2022

    2022 arXiv

  10. [15]

    When ChatGPT meets smart contract vulnerability detection: How far are we?,

    C. Chen, J. Su, J. Chen, Y. Wang, T. Bi, J. Yu, Y. Wang, X. Lin, T. Chen, and Z. Zheng, “When ChatGPT meets smart contract vulnerability detection: How far are we?,” ACM Transactions on Software Engineering and Methodology, 2023

    2023

  11. [16]

    De- tecting ponzi schemes on ethereum: Towards healthier blockchain technology,

    W. Chen, Z. Zheng, J. Cui, E. Ngai, P . Zheng, and Y. Zhou, “De- tecting ponzi schemes on ethereum: Towards healthier blockchain technology,” in Proc. of ACM Web Conference (WWW), 2018

    2018

  12. [17]

    Audit- GPT: Auditing Smart Contracts with ChatGPT,

    S. Xia, S. Shao, M. He, T. Yu, L. Song, and Y. Zhang, “Audit- GPT: Auditing Smart Contracts with ChatGPT,” arXiv preprint arXiv:2404.04306, 2024

    2024 arXiv

  13. [18]

    Stochastic neural networks for cryptocurrency price prediction,

    P . Jay, V . Kalariya, P . Parmar, S. Tanwar, N. Kumar, and M. Alazab, “Stochastic neural networks for cryptocurrency price prediction,” IEEE Access, vol. 8, pp. 82804–82818, 2020

    2020

  14. [19]

    Cryptocurrency price prediction using traditional statisti- cal and machine-learning techniques: A survey,

    A. M. Khedr, I. Arif, M. El-Bannany, S. M. Alhashmi, and M. Sreed- haran, “Cryptocurrency price prediction using traditional statisti- cal and machine-learning techniques: A survey,”Intelligent Systems in Accounting, Finance and Management , vol. 28, no. 1, pp. 3–34, 2021

    2021

  15. [20]

    Cryptocurrency price prediction using tweet volumes and sentiment analysis,

    J. Abraham, D. Higdon, J. Nelson, and J. Ibarra, “Cryptocurrency price prediction using tweet volumes and sentiment analysis,” SMU Data Science Review, vol. 1, no. 3, p. 1, 2018

    2018

  16. [21]

    DL-GuesS: Deep learning and sentiment analysis-based cryptocurrency price prediction,

    R. Parekh, N. P . Patel, N. Thakkar, R. Gupta, S. Tanwar, G. Sharma, I. E. Davidson, and R. Sharma, “DL-GuesS: Deep learning and sentiment analysis-based cryptocurrency price prediction,” IEEE Access, vol. 10, pp. 35398–35409, 2022

    2022

  17. [22]

    The economic value of NFT: Evidence from a portfolio analysis using mean–variance framework,

    H. Ko, B. Son, Y. Lee, H. Jang, and J. Lee, “The economic value of NFT: Evidence from a portfolio analysis using mean–variance framework,” Finance Research Letters, vol. 47, p. 102784, 2022

    2022

  18. [23]

    Mapping the NFT revolution: market trends, trade networks, and visual features,

    M. Nadini, L. Alessandretti, F. Di Giacinto, M. Martino, L. M. Aiello, and A. Baronchelli, “Mapping the NFT revolution: market trends, trade networks, and visual features,” Scientific Reports , vol. 11, no. 1, p. 20902, 2021

    2021

  19. [24]

    A detailed and real-time performance monitoring framework for blockchain systems,

    P . Zheng, Z. Zheng, X. Luo, X. Chen, and X. Liu, “A detailed and real-time performance monitoring framework for blockchain systems,” in Proc. of IEEE/ACM International Conference on Software Engineering (ICSE), May 2018

    2018

  20. [25]

    Anomaly detection based on traffic monitoring for secure blockchain networking,

    J. Kim, M. Nakashima, W. Fan, S. Wuthier, X. Zhou, I. Kim, and S.-Y. Chang, “Anomaly detection based on traffic monitoring for secure blockchain networking,” in Proc. of IEEE International Conference on Blockchain and Cryptocurrency (ICBC) , pp. 1–9, May 2021

    2021

  21. [26]

    A systematic literature review on blockchain governance,

    Y. Liu, Q. Lu, L. Zhu, H.-Y. Paik, and M. Staples, “A systematic literature review on blockchain governance,” arXiv [cs.SE] , May 2021

    2021

  22. [27]

    A traceability analysis of Monero’s blockchain,

    A. Kumar, C. Fischer, S. Tople, and P . Saxena, “A traceability analysis of Monero’s blockchain,” in Proc. of European Symposium on Research in Computer Security (ESORICS), pp. 153–173, Springer, 2017

    2017

  23. [28]

    An empirical analysis of traceability in the Monero blockchain,

    M. M ¨oser, K. Soska, E. Heilman, K. Lee, H. Heffan, S. Srivastava, K. Hogan, J. Hennessey, A. Miller, A. Narayanan, et al. , “An empirical analysis of traceability in the Monero blockchain,” arXiv preprint arXiv:1704.04299, 2017

    2017 arXiv

  24. [29]

    An empirical analysis of anonymity in Zcash,

    G. Kappos, H. Yousaf, M. Maller, and S. Meiklejohn, “An empirical analysis of anonymity in Zcash,” in Proc. of USENIX Security Symposium (USENIX Security), pp. 463–477, 2018

    2018

  25. [30]

    The unreasonable effectiveness of address clustering,

    M. Harrigan and C. Fretter, “The unreasonable effectiveness of address clustering,” in Proc. of IEEE International Conference on Advanced and Trusted Computing (ATC), pp. 368–373, 2016

    2016

  26. [31]

    Address clustering heuristics for Ethereum,

    F. Victor, “Address clustering heuristics for Ethereum,” in Proc. of International Conference on Financial Cryptography and Data Security (FC), pp. 617–633, Springer, 2020

    2020

  27. [32]

    Demystifying bitcoin address behavior via graph neural networks,

    Z. Huang, Y. Huang, P . Qian, J. Chen, and Q. He, “Demystifying bitcoin address behavior via graph neural networks,” in Proc. of IEEE International Conference on Data Engineering (ICDE), pp. 1747– 1760, IEEE, 2023

    2023

  28. [33]

    Support-vector networks,

    C. Cortes and V . Vapnik, “Support-vector networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, 1995

    1995

  29. [34]

    Random Forests,

    L. Breiman, “Random Forests,” Machine Learning , vol. 45, no. 1, pp. 5–32, 2001

    2001

  30. [35]

    XGBoost: A scalable tree boosting system,

    T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. of ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 785–794, ACM, 2016

    2016

  31. [36]

    Learning repre- sentations by back-propagating errors,

    D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning repre- sentations by back-propagating errors,” Nature, vol. 323, no. 6088, pp. 533–536, 1986

    1986

  32. [37]

    Fraud detection in blockchains using machine learning,

    B. Kilic, A. Sen, and C. Ozturan, “Fraud detection in blockchains using machine learning,” in Proc. of IEEE International Conference on Blockchain Computing and Applications (BCCA) , pp. 214–218, Sept. 2022

    2022

  33. [38]

    Detecting fraudulent accounts on blockchain: A supervised approach,

    M. Ostapowicz and K. ˙Zbikowski, “Detecting fraudulent accounts on blockchain: A supervised approach,” in Proc. of International Conference on Web Information Systems Engineering (WISE) , pp. 18– 31, Springer, 2019

    2019

  34. [39]

    Time series analysis for bitcoin transactions: The case of pirate@ 40’s HYIP scheme,

    K. Toyoda, T. Ohtsuki, and P . Takis Mathiopoulos, “Time series analysis for bitcoin transactions: The case of pirate@ 40’s HYIP scheme,” in IEEE International Conference on Data Mining Workshops (ICDMW), pp. 151–155, Nov. 2018

    2018

  35. [40]

    D. C. Montgomery and G. C. Runger, Applied Statistics and Proba- bility for Engineers. Wiley, 2019

    2019

  36. [41]

    Proximity-based anomaly detection using sparse structure learning,

    T. Id ´e, A. C. Lozano, N. Abe, and Y. Liu, “Proximity-based anomaly detection using sparse structure learning,” in Proc. of SIAM International Conference on Data Mining (SDM) , pp. 97–108, 2009

    2009

  37. [42]

    Isolation forest,

    F. T. Liu, K. M. Ting, and Z.-H. Zhou, “Isolation forest,” in Proc. of IEEE International Conference on Data Mining (ICDM) , pp. 413–422, 2008

    2008

  38. [43]

    Estimating the support of a high-dimensional distri- bution,

    B. Sch ¨olkopf, J. C. Platt, J. Shawe-Taylor, A. J. Smola, and R. C. Williamson, “Estimating the support of a high-dimensional distri- bution,” Neural Computation, vol. 13, no. 7, pp. 1443–1471, 2001

    2001

  39. [44]

    Blockchain net- works: Data structures of Bitcoin, Monero, Zcash, Ethereum, Rip- ple, and IoTA,

    C. G. Akcora, Y. R. Gel, and M. Kantarcioglu, “Blockchain net- works: Data structures of Bitcoin, Monero, Zcash, Ethereum, Rip- ple, and IoTA,” Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, vol. 12, no. 1, p. e1436, 2022

    2022

  40. [45]

    Graph analysis of the Ethereum blockchain data: A survey of datasets, methods, and future work,

    A. Khan, “Graph analysis of the Ethereum blockchain data: A survey of datasets, methods, and future work,” in Proc. of IEEE International Conference on Blockchain (Blockchain) , pp. 250–257, 2022

    2022

  41. [46]

    Fraud detection in ethereum using web-scraping and natural language processing techniques,

    Z. Benetti, “Fraud detection in ethereum using web-scraping and natural language processing techniques,” Master’s thesis, ETH Zurich, 2021

    2021

  42. [47]

    Formal verification of smart contracts: Short paper,

    K. Bhargavan, A. Delignat-Lavaud, C. Fournet, A. Gollamudi, G. Gonthier, N. Kobeissi, N. Kulatova, A. Rastogi, T. Sibut-Pinote, N. Swamy, et al. , “Formal verification of smart contracts: Short paper,” in Proc. of ACM Workshop on Programming Languages and Analysis for Security...

    2016

  43. [48]

    sverify: Verifying smart contracts through lazy annotation and learning,

    B. Gao, L. Shi, J. Li, J. Chang, J. Sun, and Z. Yang, “sverify: Verifying smart contracts through lazy annotation and learning,” in Proc. of International Symposium on Leveraging Applications of Formal Methods (ISoLA), pp. 453–469, Springer, 2021

    2021

  44. [49]

    Blockchain explorer: An analytical process and investigation environment for Bitcoin,

    H. Kuzuno and C. Karam, “Blockchain explorer: An analytical process and investigation environment for Bitcoin,” in Proc. of APWG Symposium on Electronic Crime Research (eCrime) , pp. 9–16, 2017

    2017

  45. [50]

    BitExtract: Interactive visualization for extracting bitcoin exchange intelligence,

    X. Yue, X. Shu, X. Zhu, X. Du, Z. Yu, D. Papadopoulos, and S. Liu, “BitExtract: Interactive visualization for extracting bitcoin exchange intelligence,” IEEE Transactions on Visualization and Com- puter Graphics, vol. 25, no. 1, pp. 162–171, 2018. 13

    2018

  46. [51]

    Visual- izing and analyzing entity activity on the bitcoin network,

    C. Kinkeldey, J.-D. Fekete, T. Blascheck, and P . Isenberg, “Visual- izing and analyzing entity activity on the bitcoin network,” arXiv [cs.HC], Dec. 2019

    2019

  47. [52]

    Position: What can large language models tell us about time series analysis,

    M. Jin, Y. Zhang, W. Chen, K. Zhang, Y. Liang, B. Yang, J. Wang, S. Pan, and Q. Wen, “Position: What can large language models tell us about time series analysis,” arXiv [cs.LG], Feb. 2024

    2024

  48. [53]

    Graph machine learning in the era of large language models (LLMs),

    W. Fan, S. Wang, J. Huang, Z. Chen, Y. Song, W. Tang, H. Mao, H. Liu, X. Liu, D. Yin, and Q. Li, “Graph machine learning in the era of large language models (LLMs),” arXiv [cs.LG], Apr. 2024

    2024

  49. [54]

    Exploring the potential of large language models (LLMs) in learning on graphs,

    Z. Chen, H. Mao, H. Li, W. Jin, H. Wen, X. Wei, S. Wang, D. Yin, W. Fan, H. Liu, and J. Tang, “Exploring the potential of large language models (LLMs) in learning on graphs,” arXiv [cs.LG] , July 2023

    2023

  50. [55]

    BC4LLM: A perspective of trusted artificial intelligence when blockchain meets large lan- guage models,

    H. Luo, J. Luo, and A. V . Vasilakos, “BC4LLM: A perspective of trusted artificial intelligence when blockchain meets large lan- guage models,” Neurocomputing, vol. 599, p. 128089, 2024

    2024

  51. [56]

    Blockchain for large language model security and safety: A holistic survey,

    C. Geren, A. Board, G. G. Dagher, T. Andersen, and J. Zhuang, “Blockchain for large language model security and safety: A holistic survey,” arXiv preprint arXiv:2407.20181, 2024

    2024 arXiv

  52. [57]

    Blockchain and artificial intelligence: Synergies and conflicts,

    L. Witt, A. T. Fortes, K. Toyoda, W. Samek, and D. Li, “Blockchain and artificial intelligence: Synergies and conflicts,” arXiv preprint arXiv:2405.13462, 2024

    2024 arXiv

  53. [58]

    Attention is all you need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” in Advances in Neural Information Processing Systems, pp. 5998–6008, 2017

    2017

  54. [59]

    Large language models are zero-shot reasoners,

    T. Kojima, S. S. Gu, M. Reid, Y. Matsuo, and Y. Iwasawa, “Large language models are zero-shot reasoners,” Advances in Neural Information Processing Systems (NeurIPS) , vol. 35, pp. 22199–22213, 2022

    2022

  55. [60]

    Exploring patterns in LLM integration-a study on architectural considerations and design pat- terns in LLM dependent applications,

    S. Ganesh and R. Sahlqvist, “Exploring patterns in LLM integration-a study on architectural considerations and design pat- terns in LLM dependent applications,” Master’s thesis, Chalmers University of Technology, 2024

    2024

  56. [61]

    BlockFound: Cus- tomized blockchain foundation model for anomaly detection,

    J. Yu, X. Wu, H. Liu, W. Guo, and X. Xing, “BlockFound: Cus- tomized blockchain foundation model for anomaly detection,” Oct. 2024. arXiv:2410.04039 [cs]

    2024

  57. [62]

    Enhancing zero-shot crypto sentiment with fine-tuned language model and prompt engineering,

    R. S. Wahidur, I. Tashdeed, M. Kaur, and H.-N. Lee, “Enhancing zero-shot crypto sentiment with fine-tuned language model and prompt engineering,” IEEE Access, 2024

    2024

  58. [63]

    Prop- ertyGPT: LLM-driven Formal Verification of Smart Contracts through Retrieval-Augmented Property Generation,

    Y. Liu, Y. Xue, D. Wu, Y. Sun, Y. Li, M. Shi, and Y. Liu, “Prop- ertyGPT: LLM-driven Formal Verification of Smart Contracts through Retrieval-Augmented Property Generation,” May 2024

    2024

  59. [64]

    Sligpt: A Large Language Model-Based Approach for Data Dependency Analysis on Solidity Smart Con- tracts,

    X. Ren and Q. Wei, “Sligpt: A Large Language Model-Based Approach for Data Dependency Analysis on Solidity Smart Con- tracts,” Software, vol. 3, no. 3, pp. 345–367, 2024

    2024

  60. [65]

    Blockchain Large Language Models,

    Y. Gai, L. Zhou, K. Qin, D. Song, and A. Gervais, “Blockchain Large Language Models,” Apr. 2023

    2023

  61. [68]

    Llmsmartsec: Smart contract security auditing with llm and annotated control flow graph,

    V . Mothukuri, R. M. Parizi, and J. L. Massa, “Llmsmartsec: Smart contract security auditing with llm and annotated control flow graph,” in Proc. of IEEE International Conference on Blockchain (Block- chain), pp. 434–441, 2024

    2024

  62. [69]

    CryptoTrade: A Reflective LLM-based Agent to Guide Zero-shot Cryptocurrency Trading,

    Y. Li, B. Luo, Q. Wang, N. Chen, X. Liu, and B. He, “CryptoTrade: A Reflective LLM-based Agent to Guide Zero-shot Cryptocurrency Trading,” in Proc. of Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1094–1106, 2024

    2024

  63. [70]

    ZipZap: Efficient Training of Language Models for Large-Scale Fraud De- tection on Blockchain,

    S. Hu, T. Huang, K.-H. Chow, W. Wei, Y. Wu, and L. Liu, “ZipZap: Efficient Training of Language Models for Large-Scale Fraud De- tection on Blockchain,” in Proc. of ACM Web Conference (WWW) , p. 2807–2816, 2024

    2024

  64. [71]

    Large language models for blockchain security: A systematic literature review,

    Z. He, Z. Li, S. Yang, A. Qiao, X. Zhang, X. Luo, and T. Chen, “Large language models for blockchain security: A systematic literature review,” arXiv [cs.CR], Mar. 2024

    2024

  65. [73]

    LLMs and NLP Models in Cryptocurrency Sentiment Analysis: A Com- parative Classification Study,

    K. I. Roumeliotis, N. D. Tselikas, and D. K. Nasiopoulos, “LLMs and NLP Models in Cryptocurrency Sentiment Analysis: A Com- parative Classification Study,” Big Data and Cognitive Computing , vol. 8, no. 6, 2024

    2024

  66. [74]

    Transformer-based approach for Ethereum Price Prediction Using Crosscurrency correlation and Sentiment Analysis,

    S. Singh and M. Bhat, “Transformer-based approach for Ethereum Price Prediction Using Crosscurrency correlation and Sentiment Analysis,” Jan. 2024. arXiv:2401.08077 [cs, q-fin]

    2024 arXiv

  67. [75]

    AI alignment: A comprehensive survey,

    J. Ji, T. Qiu, B. Chen, B. Zhang, H. Lou, K. Wang, Y. Duan, Z. He, J. Zhou, Z. Zhang, F. Zeng, K. Y. Ng, J. Dai, X. Pan, A. O’Gara, Y. Lei, H. Xu, B. Tse, J. Fu, S. McAleer, Y. Yang, Y. Wang, S.-C. Zhu, Y. Guo, and W. Gao, “AI alignment: A comprehensive survey,” arXiv [cs.AI],...

    2023

  68. [76]

    Unveiling the Potential of ChatGPT in Detecting Machine Unauditable Bugs in Smart Contracts: A Preliminary Evaluation and Categorization,

    B. Gao, Q. Wei, Y. Liu, and R. S. M. Goh, “Unveiling the Potential of ChatGPT in Detecting Machine Unauditable Bugs in Smart Contracts: A Preliminary Evaluation and Categorization,” in Proc. of IEEE Conference on Artificial Intelligence (CAI) , pp. 1481–1486, 2024. APPENDIX A ...

    2024

  69. [77]

    The transaction involves a sender transferring 10 ETH to a receiver.,→

  70. [78]

    The receiver address matches a known exchange wallet.,→

  71. [79]

    The transaction value is within typical limits for exchange deposits. ,→ ,→

  72. [80]

    Transaction ID

    No unusual activity or flags are associated with the transaction.,→ Decision: Non-Fraudulent Explanation: The receiver is a known exchange address, and the transaction value is reasonable. ,→ ,→ --- ### Example 2 ### Transaction Data: 14 "Transaction ID": "tx67890", "Sender Ad...

    2024

  73. [81]

    The transaction involves a sender transferring 500 ETH to a receiver.,→

  74. [82]

    The receiver address is unknown and not associated with a verified entity. ,→ ,→

  75. [83]

    The transaction value is exceptionally large compared to average transactions. ,→ ,→

  76. [84]

    Transaction ID

    The notes indicate unusual activity, further raising suspicion.,→ Decision: Fraudulent Explanation: The transaction is flagged due to the high value and the unknown receiver, which suggests potential fraud. ,→ ,→ ,→ --- ### Query ### Transaction Data: "Transaction ID": "tx5432...

    2024

  77. [85]

    - Node 2: -

    Graph Reasoning: - Node 1: - ... - Node 2: - ... - Node 3: - ... Relationships: - Edge between Node 1 and Node 2: - ... - Edge between Node 2 and Node 3: - ... Evaluation: ... Decision: Explanation:

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.