Pith. sign in

REVIEW 2 major objections 5 minor 1 cited by

Following Devils' Footprint: Towards Real-time Detection of Price Manipulation Attacks

T0 review · 2 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read Catching price-manipulation bots before they strike

desk verdict Solid engineering, real deployment evidence, but the recall number likely overstates generalization because the rules and ground truth come from the same incidents. read the letter →

arxiv 2502.03718 v1 pith:PHK7TMND submitted 2025-02-06 cs.CR

Bosi Zhang , Ningyu He , Xiaohui Hu , Kai Ma , Haoyu Wang This is my paper · ORCID
classification cs.CR
keywords pricemanipulationDeFismartcontractsecuritybytecodeanalysisstatictokenflowgraphflashloanreal-timedetection
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

This paper tries to establish that price manipulation attacks can be detected before they execute, by reading only the deployed bytecode of the attacker's contract. Existing defenses react after a transaction drains a pool; this approach looks for tell-tale token-flow patterns in the attacker's own contract the moment it appears on a chain. If the claim is right, alerts can arrive within seconds of deployment rather than after funds are lost. The authors report 91.6% recall and near-perfect precision on a hand-verified set of 84 known attacks, and they identify 616 attack contracts in the wild, only 19 of which had been publicly reported.

What carries the argument

The central object is the token flow graph (TFG): a graph whose nodes are function calls annotated with five token actions (transfer, swap, add liquidity, remove liquidity, flash loan) and whose edges are control- or data-flow dependencies recovered from bytecode by a heuristic argument-recovery algorithm. The detector restricts cross-contract analysis to so-called sensitive paths, which correspond to the fund-preparation, token-exchange, and fund-transfer stages of a DeFi attack, and then matches four formal rules on paths through the TFG.

What would settle it

Take a known two-transaction price manipulation, such as the stake-and-harvest cases the paper itself reports missing, run SMARTCAT against the deployment bytecode of the attacker contract, and check that no alarm is raised before the second transaction executes; even one such case demonstrates the gap in coverage.

Watch

Extended reading notes

Core claim

A static analyzer can identify attack contracts at deployment time by modeling the control- and data-flow of token-related function calls into a token flow graph, then matching formal rules that capture pump-and-dump, direct price manipulation, and indirect price manipulation. On a ground-truth set of 84 known attacks and 8,000 benign contracts, it finds 77 attacks with only two false positives, and in live deployment on Ethereum and Binance Smart Chain it raised 14 alarms on average 99 seconds after the attack contract appeared, with three of those attacks going on to steal over $641,000.

Load-bearing premise

The detector assumes attacks stick to the token-flow pattern of two swaps on the same pool with an interleaved transfer or liquidity action, with flash loans for direct manipulation, so attackers who use multi-transaction flows or custom DEX functions will not be flagged.

Editorial extensions

If this is right

  • If the claim is right, security teams can monitor newly deployed contracts and warn affected protocols while the attacker's contract is still dormant, before any pool is drained.
  • The 616 in-the-wild hits suggest the real volume of price manipulation is far higher than publicly reported incidents; deployment-time scanning would surface these before profits are taken.
  • The rule-based core is agnostic to the specific victim contract, so new victims using the same manipulation pattern are covered without per-project rule engineering.
  • Alarm latency of roughly 99 seconds means only attacks that deploy and strike within that window, or that split the attack across multiple transactions, would escape the pre-attack net.

Reading between the lines

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

  • The attack-pattern rules could be translated into a white-list check for DEX integration: any contract that takes a flash loan and then swaps twice on one pool with an interleaved transfer is worth blocking outright when false alarms are tolerable.
  • The structural fingerprint defined by the TFG could support anomaly scoring rather than fixed rules, potentially catching novel manipulation variants that do not exactly match the four formal patterns.
  • A testable extension is to run the same pipeline on other chains or on alternative DEX designs, such as concentrated-liquidity pools, to see whether the two-swap-on-one-pool pattern generalizes or needs additional rules.
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

2 major / 5 minor

Summary. The paper proposes SMARTCAT, a static analyzer that identifies price manipulation attack contracts from EVM bytecode alone, with the goal of raising alarms in the pre-attack stage before an exploit is launched. The system decompiles bytecode with Gigahorse, extracts intra- and inter-contract call information, builds a token flow graph (TFG), and applies four formal rules (Pump-and-Dump, Direct Price Manipulation, and Indirect Price Manipulation) to flag suspicious contracts. The evaluation reports 91.6% recall and ~100% precision on a ground-truth set of 84 attack contracts and 8,000 benign contracts, an in-the-wild scan of over 770K contracts that yields 616 candidate attack contracts, and a 50-day live deployment on Ethereum and BSC that raised 14 alarms an average of 99 seconds after deployment.

Significance. If the recall and precision figures are accepted, SMARTCAT is a meaningful advance: it is, to the authors' knowledge, the first bytecode-level detector that targets the pre-attack stage, and it ships a released artifact and ground-truth dataset. The ablation study isolating the argument recovery and sensitive path filtering modules is a strength, as is the live deployment showing that detection can complete within seconds of contract deployment. The central limitation is that the formal rules in Section 4.3.2 were derived from known incidents that also populate the evaluation set, so the headline recall may overestimate out-of-sample performance. The baseline comparison also mixes different test sets and detection targets. These issues are load-bearing for the paper's main claims but are addressable in revision.

major comments (2)
  1. [§4.3.2, §6.1, Table 6] The reported 91.6% recall on DG1 is vulnerable to training-set circularity. The detecting rules in Section 4.3.2 are clearly motivated by the ULME example in Section 3.1 and by the ElephantStatus and Cheese Bank examples in Figure 1, and all three incidents appear in the DG1 evaluation set (cases #34, #77, and #6 in Table 6). The paper does not report any temporal or held-out split, so the rules may simply encode the token-flow patterns of the very attacks used to design them. Please provide an out-of-sample evaluation, e.g., a temporal holdout of incidents reported after the rule-freeze date, or a random split where the rules are fixed before evaluation, and report recall on the held-out subset separately. Without this, the 91.6% recall should not be presented as a general detection rate.
  2. [§5, §6.1, Table 1] The baseline comparison is not apples-to-apples. DeFiRanger and FlashSyn are closed-source, so their recall values (54.8% and 42.9%) are taken from their own papers and computed over their own incident sets of 42 and 21 incidents, respectively, while SMARTCAT is evaluated on all 84 incidents in DG1. The denominators are different, so the percentage recall values are not directly comparable. Additionally, DeFiTainter detects vulnerable protocols rather than attack contracts, so the detection targets are different. The claim that SMARTCAT 'significantly outperforms' baselines is therefore not supported by the current table. Please either run the baselines on the same DG1 incidents (where artifacts permit) or report a matched-subset comparison with identical denominators, and explicitly state the different detection targets in the comparison.
minor comments (5)
  1. [§4.3.2, Rule 2] The first argument of FL in Rule 2 appears to denote a token, but the flashloan semantics in Figure 5 define FL(pr, token, amt, to) with the lending protocol first. Please align the notation so the rule's four arguments match the definition.
  2. [§10] In the Related Work section, 'DeFiRange' should be spelled 'DeFiRanger' to match the tool name used everywhere else.
  3. [Abstract, §7] The phrase 'accounting for $9.25M financial losses' refers to the 166 successful attack contracts, not to all 616 identified contracts. Consider clarifying this in the abstract to avoid overstatement.
  4. [§6.1] The statistic that 68 out of 84 cases (81.0%) are alerted within the attack window is not shown in Table 1. Please add a column or state explicitly how this number is derived from the TP/FN counts.
  5. [§8, Table 3 and Table 4] The '99 seconds after deployment on average' figure mixes three successful attacks, four failed attempts, and seven unfinished attempts. Please state the average for the successful cases alone, as that is the most relevant for the timeliness claim.

Circularity Check

1 steps flagged · score 2.0 of 10

Detection rules are hand-built from the same known incidents that populate DG1, so the 91.6% recall is partly a train-on-test score; no parameters are fitted and the real-time/profit evidence is independent, so circularity is mild.

  1. other [Section 4.3.2 (rules); Section 5 (DG1 construction); Section 2.2/3.1 (motivating examples); Appendix Table 6]
    "we formalize a set of rules to characterize two types of price manipulation attack behaviors ... DG1 comprises 84 attack events labeled as price manipulation, sourced from various mainstream platforms [17,26,27], the publicly released datasets of FlashSyn [36] and DeFiRanger [71] ... Figure 1(a) illustrates a real-world example of direct price manipulation, where ElephantStatus was exploited ... Figure 1(b) illustrates a concrete example of an indirect price manipulation, which occurred on Nov. 6th, 2020 ... ULME project has been attacked in a DPM way on Oct. 25th, 2022"

    The four rules in §4.3.2 are hand-written formalizations of the DPM/IPM behaviors introduced in §2.2, which are illustrated with Cheese Bank, ElephantStatus, and ULME. These exact incidents appear in DG1 (Table 6: cases #6, #34, #77), and DG1 is the dataset used for the RQ1 recall of 91.6%. No temporal or held-out split is reported, and the paper does not state that the rule set was frozen before the 84 incidents were collected. The recall therefore measures whether the rules match the incident reports that motivated them; it is a training-set score rather than an out-of-sample estimate. The §7 profit-validated detections and §8 real-time alarms use independent transaction evidence and do not share this circularity.

full rationale

SMARTCAT's derivation chain is mostly self-contained: token-flow semantics are recovered from bytecode via Gigahorse and hand-written heuristics; the detecting rules are explicit and transparent; and the large-scale (RQ2) and real-time (RQ3) evaluations rest on independent on-chain evidence (attack transactions, profits, labels) rather than on the rules' own outputs. No parameter is fitted to maximize the reported metrics, and no load-bearing claim is justified solely by a self-citation. The one genuine circularity-adjacent issue is that the rule set was designed from the same publicly reported incidents that populate DG1, so the headline 91.6% recall is partly a check that the rules match their motivating examples. This is an evaluation-leakage concern, not an equivalence-by-construction; it is mitigated by the in-the-wild and real-time validation, so the overall circularity score is low.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

SMARTCAT introduces no new physical entities or fitted numerical parameters. Its central scaffolding is a set of hand-crafted heuristics and pattern rules (argument recovery, sensitive path filtering, detection rules) plus tool-dependent assumptions about Gigahorse. The performance claims depend on these heuristics.

free parameters (3)
  • cross_contract_analysis_depth = 3
    Hand-chosen limit on recursive cross-contract analysis depth, set to three in the evaluation (§5).
  • gigahorse_timeout = 120 seconds
    Timeout for Gigahorse decompilation, chosen for scalability (§5).
  • benign_contracts_top_k = 8,000
    Number of highest-transaction-volume contracts treated as benign ground truth (§5).
assumptions (5)
  • domain assumption Attack contracts embed all attack logic in a single contract and launch the attack in one transaction.
    Stated in §3.2 and §9 as a design choice; authors admit multi-transaction attacks are not handled.
  • ad hoc to paper Price manipulation necessarily involves a pump-and-dump pattern of two swaps on the same pool, with a transfer/add/remove liquidity action in between (Rules 1-4).
    These are the hand-crafted detection rules in §4.3.2; they define the attack space rather than being derived from first principles.
  • domain assumption The 8,000 contracts with the highest transaction volume are non-malicious.
    Used to construct DG2 benign set; possible selection bias (§5).
  • domain assumption Gigahorse decompilation is accurate for all relevant contracts.
    SMARTCAT is built on Gigahorse; false negatives are attributed to Gigahorse failures in §6.1.
  • ad hoc to paper Callee address and function selector recovery heuristics are correct.
    Address extraction from Memory/Storage/calldata and nearest-PUSH4 selector recovery (§4.1.1) are heuristic and can fail on obfuscated code.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Following Devils' Footprint: Towards Real-time Detection of Price Manipulation Attacks." pith.science (2026). https://pith.science/paper/PHK7TMND

@misc{pith2026250203718,
  author       = {Pith},
  title        = {Pith review of: Following Devils' Footprint: Towards Real-time Detection of Price Manipulation Attacks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PHK7TMND}},
  note         = {Machine review of arXiv:2502.03718}
}
abstract

Price manipulation attack is one of the notorious threats in decentralized finance (DeFi) applications, which allows attackers to exchange tokens at an extensively deviated price from the market. Existing efforts usually rely on reactive methods to identify such kind of attacks after they have happened, e.g., detecting attack transactions in the post-attack stage, which cannot mitigate or prevent price manipulation attacks timely. From the perspective of attackers, they usually need to deploy attack contracts in the pre-attack stage. Thus, if we can identify these attack contracts in a proactive manner, we can raise alarms and mitigate the threats. With the core idea in mind, in this work, we shift our attention from the victims to the attackers. Specifically, we propose SMARTCAT, a novel approach for identifying price manipulation attacks in the pre-attack stage proactively. For generality, it conducts analysis on bytecode and does not require any source code and transaction data. For accuracy, it depicts the control- and data-flow dependency relationships among function calls into a token flow graph. For scalability, it filters out those suspicious paths, in which it conducts inter-contract analysis as necessary. To this end, SMARTCAT can pinpoint attacks in real time once they have been deployed on a chain. The evaluation results illustrate that SMARTCAT significantly outperforms existing baselines with 91.6% recall and ~100% precision. Moreover, SMARTCAT also uncovers 616 attack contracts in-the-wild, accounting for \$9.25M financial losses, with only 19 cases publicly reported. By applying SMARTCAT as a real-time detector in Ethereum and Binance Smart Chain, it has raised 14 alarms 99 seconds after the corresponding deployment on average. These attacks have already led to $641K financial losses, and seven of them are still waiting for their ripe time.

Figures

Figures reproduced from arXiv: 2502.03718 by the authors.

Figure 1
Figure 1. Two types of price manipulation. 2.2 Price Manipulation Price manipulation in DeFi is basically achieved by exploiting how DEXes calculate token prices. Typically, for a token in a liquidity pool, the lower the supply, the higher the price. When the supply balance between tokens is disrupted, price deviations may occur. The existence of flashloan further wors￾ens this issue by enabling someone to borrow large amount… view at source ↗
Figure 2
Figure 2. Workflow and architecture of SMARTCAT. tween contracts can lead to path explosion during symbolic execution. As shown in Listing 1, recovering and identifying semantics at both intra- and inter-contract function calls is crucial for accurately determining the contract’s behavior. Challenge 2: Scalability issue. Detecting price manipula￾tion attack contracts requires exploring the paths correspond￾ing to conducting a… view at source ↗
Figure 3
Figure 3. Extract function signature by Gigahorse and our [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Heuristically infer potential token swap actions. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The definition of token action related semantics. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: The xFCG and TFG of the attack contract in the [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 9
Figure 9. Figure 9: The relationship between the consumed time and [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 7
Figure 7. Figure 7: Distribution of attack profits by deployment time. [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 8
Figure 8. Figure 8: Distribution of contract runtimes. ally, [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Detecting State Manipulation Vulnerabilities in Smart Contracts Using LLM and Static Analysis

    cs.SE 2025-06 conditional novelty 4.0 of 10

    A pipeline that uses LLM reasoning over dependency and propagation graphs detects price manipulation vulnerabilities in four real DeFi attacks in under one minute.

Reference graph

Works this paper leans on

80 extracted references · 73 canonical work pages · cited by 1 Pith paper

  1. [1]

    https://eips.ethereum.org/ EIPS/eip-20, 2015

    Erc-20 token standard. https://eips.ethereum.org/ EIPS/eip-20, 2015

    work page 2015

  2. [2]

    https: //eips.ethereum.org/EIPS/eip-1155, 2018

    Erc-1155 multi token standard. https: //eips.ethereum.org/EIPS/eip-1155, 2018

    work page 2018

  3. [3]

    https:// eips.ethereum.org/EIPS/eip-721, 2018

    Erc-721 non-fungible token standard. https:// eips.ethereum.org/EIPS/eip-721, 2018

    work page 2018

  4. [4]

    https://github.com/ConsenSys/mythril, 2018

    Mythril: Security analysis tool for ethereum smart con- tracts. https://github.com/ConsenSys/mythril, 2018

    work page 2018

  5. [5]

    https://eips.ethereum.org/EIPS/eip- 1967, 2019

    Eip-1967. https://eips.ethereum.org/EIPS/eip- 1967, 2019

    work page 1967

  6. [6]

    https:// peckshield.medium.com/cheese-bank-incident- root-cause-analysis-d076bf87a1e7 , 2020

    The cheese bank incident. https:// peckshield.medium.com/cheese-bank-incident- root-cause-analysis-d076bf87a1e7 , 2020

    work page 2020

  7. [7]

    https://github.com/xf97/BiAn/, 2021

    work page 2021

  8. [8]

    The collapse of ftx.https://www.investopedia.com/ what-went-wrong-with-ftx-6828447 , 2023

    work page 2023

Show all 80 references
  1. [9]

    https://x.com/ Phalcon_xyz/status/1732354930529435940, 2023

    The elephantstatus incident. https://x.com/ Phalcon_xyz/status/1732354930529435940, 2023

    2023

  2. [10]

    https://github.com/ tintinweb/smart-contract-sanctuary, 2023

    Smart contract sanctuary. https://github.com/ tintinweb/smart-contract-sanctuary, 2023

    2023

  3. [11]

    https://aave.com, 2024

    Aave. https://aave.com, 2024

    2024

  4. [12]

    https: //etherscan.io/address/ 0x2F99fb66Ea797E7fA2d07262402Ab38bd5e53B12, 2024

    Attack contract example. https: //etherscan.io/address/ 0x2F99fb66Ea797E7fA2d07262402Ab38bd5e53B12, 2024

    2024

  5. [13]

    https://www.binance.com/en/square/ post/8851640245434, 2024

    Biance. https://www.binance.com/en/square/ post/8851640245434, 2024

    2024

  6. [14]

    https://www.coindesk.com, 2024

    Coindesk. https://www.coindesk.com, 2024

    2024

  7. [15]

    https://dune.com/ queries/4037888, 2024

    Daily contract deployment. https://dune.com/ queries/4037888, 2024

    2024

  8. [16]

    https://app.dedaub.com/ decompile, 2024

    Dedaub decompiler. https://app.dedaub.com/ decompile, 2024

    2024

  9. [17]

    https://github.com/SunWeb3Sec/ DeFiHackLabs, 2024

    Defihacklab. https://github.com/SunWeb3Sec/ DeFiHackLabs, 2024

    2024

  10. [18]

    https://defillama.com, 2024

    Defillama. https://defillama.com, 2024

    2024

  11. [19]

    https: //docs.etherscan.io, 2024

    Etherscan api documentation. https: //docs.etherscan.io, 2024

    2024

  12. [20]

    https://forta.org/blog/how-fortas- predictive-ml-models-detect-attacks- before-exploitation/, 2024

    Forta. https://forta.org/blog/how-fortas- predictive-ml-models-detect-attacks- before-exploitation/, 2024

    2024

  13. [21]

    https://book.getfoundry.sh/, 2024

    The foundry book. https://book.getfoundry.sh/, 2024

    2024

  14. [22]

    https://github.com/ethereum/go- ethereum, 2024

    Geth. https://github.com/ethereum/go- ethereum, 2024

    2024

  15. [23]

    https://degatchi.com/articles/mev-smart- contract-obfuscation, 2024

    Mev smart contract obfuscation techniques. https://degatchi.com/articles/mev-smart- contract-obfuscation, 2024

    2024

  16. [24]

    https: //docs.openzeppelin.com/contracts/4.x/api/ proxy, 2024

    Openzeppelin proxies document. https: //docs.openzeppelin.com/contracts/4.x/api/ proxy, 2024

    2024

  17. [25]

    https://pancakeswap.finance, 2024

    Pancakeswap. https://pancakeswap.finance, 2024

    2024

  18. [26]

    https://rekt.news, 2024

    Rekt. https://rekt.news, 2024

    2024

  19. [27]

    https://www.slowmist.com/, 2024

    Slowmist. https://www.slowmist.com/, 2024. 15

    2024

  20. [28]

    https://docs.soliditylang.org/en/ v0.8.26/, 2024

    Solidity. https://docs.soliditylang.org/en/ v0.8.26/, 2024

    2024

  21. [29]

    https://tokenterminal.com/ terminal/datasets/trending-contracts, 2024

    Token terminal. https://tokenterminal.com/ terminal/datasets/trending-contracts, 2024

    2024

  22. [30]

    https://x.com/BeosinAlert/ status/1584888021299916801, 2024

    The ulme attack. https://x.com/BeosinAlert/ status/1584888021299916801, 2024

    2024

  23. [31]

    https://docs.uniswap.org, 2024

    Uniswap protocol. https://docs.uniswap.org, 2024

    2024

  24. [32]

    https://vyperlang.org/, 2024

    Vyper documentation. https://vyperlang.org/, 2024

    2024

  25. [33]

    Sailfish: Vet- ting smart contract state-inconsistency bugs in seconds

    Priyanka Bose, Dipanjan Das, Yanju Chen, Yu Feng, Christopher Kruegel, and Giovanni Vigna. Sailfish: Vet- ting smart contract state-inconsistency bugs in seconds. In 2022 IEEE Symposium on Security and Privacy(SP), pages 161–178. IEEE, 2022

    2022

  26. [34]

    Ethainter: a smart contract security analyzer for composite vulnera- bilities

    Lexi Brent, Neville Grech, Sifis Lagouvardos, Bern- hard Scholz, and Yannis Smaragdakis. Ethainter: a smart contract security analyzer for composite vulnera- bilities. In Proceedings of the 41st ACM SIGPLAN Conference on Programming Language Design and Implementation, pages 45...

    2020

  27. [35]

    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

  28. [36]

    Flashsyn: Flash loan attack synthesis via counter ex- ample driven approximation

    Zhiyang Chen, Sidi Mohamed Beillahi, and Fan Long. Flashsyn: Flash loan attack synthesis via counter ex- ample driven approximation. In Proceedings of the IEEE/ACM 46th International Conference on Software Engineering, pages 1–13, 2024

    2024

  29. [37]

    Smar- tian: Enhancing smart contract fuzzing with static and dynamic data-flow analyses

    Jaeseung Choi, Doyeon Kim, Soomin Kim, Gustavo Grieco, Alex Groce, and Sang Kil Cha. Smar- tian: Enhancing smart contract fuzzing with static and dynamic data-flow analyses. In 2021 36th IEEE/ACM International Conference on Automated Software Engineering (ASE), pages 227–239. ...

    2021

  30. [38]

    Introducing Ethereum and solidity, vol- ume 1

    Chris Dannen. Introducing Ethereum and solidity, vol- ume 1. Springer, 2017

    2017

  31. [39]

    Slither: a static analysis framework for smart con- tracts

    Josselin Feist, Gustavo Grieco, and Alex Groce. Slither: a static analysis framework for smart con- tracts. In 2019 IEEE/ACM 2nd International Workshop on Emerging Trends in Software Engineering for Blockchain (WETSEB), pages 8–15. IEEE, 2019

    2019

  32. [40]

    {ETHBMC}: A bounded model checker for smart con- tracts

    Joel Frank, Cornelius Aschermann, and Thorsten Holz. {ETHBMC}: A bounded model checker for smart con- tracts. In 29th USENIX Security Symposium (USENIX Security 20), pages 2757–2774, 2020

    2020

  33. [41]

    Gigahorse: thorough, declarative de- compilation of smart contracts

    Neville Grech, Lexi Brent, Bernhard Scholz, and Yan- nis Smaragdakis. Gigahorse: thorough, declarative de- compilation of smart contracts. In 2019 IEEE/ACM 41st International Conference on Software Engineering (ICSE), pages 1176–1186. IEEE, 2019

    2019

  34. [42]

    Mad- max: Surviving out-of-gas conditions in ethereum smart contracts

    Neville Grech, Michael Kong, Anton Jurisevic, Lexi Brent, Bernhard Scholz, and Yannis Smaragdakis. Mad- max: Surviving out-of-gas conditions in ethereum smart contracts. Proceedings of the ACM on Programming Languages, 2(OOPSLA):1–27, 2018

    2018

  35. [43]

    Elipmoc: advanced decompilation of ethereum smart contracts

    Neville Grech, Sifis Lagouvardos, Ilias Tsatiris, and Yan- nis Smaragdakis. Elipmoc: advanced decompilation of ethereum smart contracts. Proceedings of the ACM on Programming Languages, 6(OOPSLA1):1–27, 2022

    2022

  36. [44]

    Echidna: effective, usable, and fast fuzzing for smart contracts

    Gustavo Grieco, Will Song, Artur Cygan, Josselin Feist, and Alex Groce. Echidna: effective, usable, and fast fuzzing for smart contracts. In Proceedings of the 29th ACM SIGSOFT international symposium on software testing and analysis, pages 557–560, 2020

    2020

  37. [45]

    Learning to fuzz from symbolic execution with application to smart con- tracts

    Jingxuan He, Mislav Balunovi ´c, Nodar Ambroladze, Petar Tsankov, and Martin Vechev. Learning to fuzz from symbolic execution with application to smart con- tracts. In Proceedings of the 2019 ACM SIGSAC conference on computer and communications security, pages 531–548, 2019

    2019

  38. [46]

    In 30th USENIX security symposium (USENIX Security 21), pages 1271–1288, 2021

    Ningyu He, Ruiyi Zhang, Haoyu Wang, Lei Wu, Xiapu Luo, Yao Guo, Ting Yu, and Xuxian Jiang.{EOSAFE}: security analysis of {EOSIO} smart contracts. In 30th USENIX security symposium (USENIX Security 21), pages 1271–1288, 2021

    2021

  39. [47]

    Large language models for blockchain security: A systematic literature review

    Zheyuan He, Zihao Li, and Sen Yang. Large language models for blockchain security: A systematic literature review. arXiv preprint arXiv:2403.14280, 2024

    2024 arXiv

  40. [48]

    Defining the ethereum virtual machine for interactive theorem provers

    Yoichi Hirai. Defining the ethereum virtual machine for interactive theorem provers. In Financial Cryptography and Data Security: FC 2017 International Workshops, W AHC,BITCOIN, VOTING, WTSC, and TA, Sliema, Malta, April 7, 2017, Revised Selected Papers 21, pages 520–535. Spri...

    2017

  41. [49]

    Con- tractfuzzer: Fuzzing smart contracts for vulnera- bility detection

    Bo Jiang, Ye Liu, and Wing Kwong Chan. Con- tractfuzzer: Fuzzing smart contracts for vulnera- bility detection. In Proceedings of the 33rd ACM/IEEE international conference on automated software engineering, pages 259–269, 2018

    2018

  42. [50]

    Defitainter: Detecting price manipu- lation vulnerabilities in defi protocols

    Queping Kong, Jiachi Chen, Yanlin Wang, Zigui Jiang, and Zibin Zheng. Defitainter: Detecting price manipu- lation vulnerabilities in defi protocols. In Proceedings of the 32nd ACM SIGSOFT International Symposium 16 on Software Testing and Analysis, pages 1144–1156, 2023

    2023

  43. [51]

    ground- hog day

    Levi Taiji Li, Ningyu He, Haoyu Wang, and Mu Zhang. Veteos: Statically vetting eosio contracts for the “ground- hog day” vulnerabilities

  44. [52]

    Varlifter: Recov- ering variables and types from bytecode of solidity smart contracts

    Yichuan Li, Wei Song, and Jeff Huang. Varlifter: Recov- ering variables and types from bytecode of solidity smart contracts. Proceedings of the ACM on Programming Languages, 8(OOPSLA2):1–29, 2024

    2024

  45. [53]

    Making smart contracts smarter

    Loi Luu, Duc-Hiep Chu, Hrishi Olickel, Prateek Saxena, and Aquinas Hobor. Making smart contracts smarter. In Proceedings of the 2016 ACM SIGSAC conference on computer and communications security, pages 254– 269, 2016

    2016

  46. [54]

    Fuzzing: Art, science, and engineering

    Valentin JM Manes, HyungSeok Han, Choongwoo Han, Sang Kil Cha, Manuel Egele, Edward J Schwartz, and Maverick Woo. Fuzzing: Art, science, and engineering. arXiv preprint arXiv:1812.00140, 2018

    2018 arXiv

  47. [55]

    Toward automated detecting unanticipated price feed in smart contract

    Yifan Mo, Jiachi Chen, Yanlin Wang, and Zibin Zheng. Toward automated detecting unanticipated price feed in smart contract. In Proceedings of the 32nd ACM SIGSOFT International Symposium on Software Testing and Analysis, pages 1257–1268, 2023

    2023

  48. [56]

    Manticore: A user- friendly symbolic execution framework for binaries and smart contracts

    Mark Mossberg, Felipe Manzano, Eric Hennenfent, Alex Groce, Gustavo Grieco, Josselin Feist, Trent Brunson, and Artem Dinaburg. Manticore: A user- friendly symbolic execution framework for binaries and smart contracts. In 2019 34th IEEE/ACM International Conference on Automated...

    2019

  49. [57]

    Bitcoin: A peer-to-peer electronic cash system

    Satoshi Nakamoto. Bitcoin: A peer-to-peer electronic cash system. Satoshi Nakamoto, 2008

    2008

  50. [58]

    Automated generation of security-centric descriptions for smart contract bytecode

    Yu Pan, Zhichao Xu, Levi Taiji Li, Yunhe Yang, and Mu Zhang. Automated generation of security-centric descriptions for smart contract bytecode. InProceedings of the 32nd ACM SIGSOFT International Symposium on Software Testing and Analysis, pages 1244–1256, 2023

    2023

  51. [59]

    Attacking the defi ecosystem with flash loans for fun and profit

    Kaihua Qin, Liyi Zhou, Benjamin Livshits, and Arthur Gervais. Attacking the defi ecosystem with flash loans for fun and profit. In International conference on financial cryptography and data security, pages 3–32. Springer, 2021

    2021

  52. [60]

    O’Reilly Media, Inc

    Siraj Raval. Decentralized applications: harnessing Bitcoin’s blockchain technology. " O’Reilly Media, Inc.", 2016

    2016

  53. [61]

    Lookahead: Preventing defi attacks via unveiling adversarial contracts

    Shoupeng Ren, Tianyu Tu, Jian Liu, Di Wu, and Kui Ren. Lookahead: Preventing defi attacks via unveiling adversarial contracts. arXiv preprint arXiv:2401.07261, 2024

    2024 arXiv

  54. [62]

    A systematic review of decentralized finance protocols

    Kaushal Shah, Dhruvil Lathiya, Naimish Lukhi, Keyur Parmar, and Harshal Sanghvi. A systematic review of decentralized finance protocols. International Journal of Intelligent Networks, 2023

    2023

  55. [63]

    Ity- fuzz: Snapshot-based fuzzer for smart contract

    Chaofan Shou, Shangyin Tan, and Koushik Sen. Ity- fuzz: Snapshot-based fuzzer for smart contract. In Proceedings of the 32nd ACM SIGSOFT International Symposium on Software Testing and Analysis, pages 322–333, 2023

    2023

  56. [64]

    Evil under the sun: Understanding and discovering attacks on ethereum decentralized applications

    Liya Su, Xinyue Shen, Xiangyu Du, Xiaojing Liao, Xi- aoFeng Wang, Luyi Xing, and Baoxu Liu. Evil under the sun: Understanding and discovering attacks on ethereum decentralized applications. In 30th USENIX Security Symposium (USENIX Security 21), pages 1307–1324, 2021

    2021

  57. [65]

    All your tokens are belong to us: Demystifying address verification vulnerabilities in solidity smart contracts

    Tianle Sun, Ningyu He, Jiang Xiao, Yinliang Yue, Xiapu Luo, and Haoyu Wang. All your tokens are belong to us: Demystifying address verification vulnerabilities in solidity smart contracts. In The 33rd USENIX Security Symposium, 2024

    2024

  58. [66]

    Gptscan: Detecting logic vulnerabilities in smart contracts by combining gpt with program analysis

    Yuqiang Sun, Daoyuan Wu, Yue Xue, Han Liu, Haijun Wang, Zhengzi Xu, Xiaofei Xie, and Yang Liu. Gptscan: Detecting logic vulnerabilities in smart contracts by combining gpt with program analysis. In Proceedings of the IEEE/ACM 46th International Conference on Software Engineeri...

    2024

  59. [67]

    Se- curify: Practical security analysis of smart contracts

    Petar Tsankov, Andrei Dan, Dana Drachsler-Cohen, Arthur Gervais, Florian Buenzli, and Martin Vechev. Se- curify: Practical security analysis of smart contracts. In Proceedings of the 2018 ACM SIGSAC conference on computer and communications security, pages 67–82, 2018

    2018

  60. [68]

    Defiguard: A price manipulation detection service in defi using graph neural networks

    Dabao Wang, Bang Wu, Xingliang Yuan, Lei Wu, Yajin Zhou, and Helei Cui. Defiguard: A price manipulation detection service in defi using graph neural networks. arXiv preprint arXiv:2406.11157, 2024

    2024 arXiv

  61. [69]

    Towards understanding flash loan and its applications in defi ecosystem

    Dabao Wang, Siwei Wu, Ziling Lin, Lei Wu, Xingliang Yuan, Yajin Zhou, Haoyu Wang, and Kui Ren. Towards understanding flash loan and its applications in defi ecosystem. arXiv preprint arXiv:2010.12252, 2020

    2010 arXiv

  62. [70]

    Smartinv: Multimodal learning for smart contract invari- ant inference

    Sally Junsong Wang, Kexin Pei, and Junfeng Yang. Smartinv: Multimodal learning for smart contract invari- ant inference. In 2024 IEEE Symposium on Security and Privacy (SP), pages 126–126. IEEE Computer So- ciety, 2024. 17

    2024

  63. [71]

    Defiranger: Detect- ing defi price manipulation attacks

    Siwei Wu, Zhou Yu, Dabao Wang, Yajin Zhou, Lei Wu, Haoyu Wang, and Xingliang Yuan. Defiranger: Detect- ing defi price manipulation attacks. IEEE Transactions on Dependable and Secure Computing, 2023

    2023

  64. [72]

    Harvey: A greybox fuzzer for smart contracts

    Valentin Wüstholz and Maria Christakis. Harvey: A greybox fuzzer for smart contracts. In Proceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering, pages 1398– 1409, 2020

    2020

  65. [73]

    Uncover the premeditated attacks: Detecting exploitable reentrancy vulnerabilities by iden- tifying attacker contracts

    Shuo Yang, Jiachi Chen, Mingyuan Huang, Zibin Zheng, and Yuan Huang. Uncover the premeditated attacks: Detecting exploitable reentrancy vulnerabilities by iden- tifying attacker contracts. In Proceedings of the IEEE/ACM 46th International Conference on Software Engineering, pa...

    2024

  66. [74]

    Detecting state inconsistency bugs in dapps via on-chain transaction replay and fuzzing

    Mingxi Ye, Yuhong Nan, Zibin Zheng, Dongpeng Wu, and Huizhong Li. Detecting state inconsistency bugs in dapps via on-chain transaction replay and fuzzing. In Proceedings of the 32nd ACM SIGSOFT International Symposium on Software Testing and Analysis, pages 298–309, 2023

    2023

  67. [75]

    Bytecode obfuscation for smart con- tracts

    Qifan Yu, Pengcheng Zhang, Hai Dong, Yan Xiao, and Shunhui Ji. Bytecode obfuscation for smart con- tracts. In 2022 29th Asia-Pacific Software Engineering Conference (APSEC), pages 566–567. IEEE, 2022

    2022

  68. [76]

    Decentralized finance

    Dirk A Zetzsche, Douglas W Arner, and Ross P Buckley. Decentralized finance. Journal of Financial Regulation, 6(2):172–203, 2020

    2020

  69. [77]

    Bian: smart contract source code obfuscation

    Pengcheng Zhang, Qifan Yu, Yan Xiao, Hai Dong, Xi- apu Luo, Xiao Wang, and Meng Zhang. Bian: smart contract source code obfuscation. IEEE Transactions on Software Engineering, 2023

    2023

  70. [78]

    Demystifying exploitable bugs in smart contracts

    Zhuo Zhang, Brian Zhang, Wen Xu, and Zhiqiang Lin. Demystifying exploitable bugs in smart contracts. In 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE), pages 615–627. IEEE, 2023

    2023

  71. [79]

    Sok: Decentralized finance (defi) attacks

    Liyi Zhou, Xihan Xiong, Jens Ernstberger, Stefanos Chaliasos, Zhipeng Wang, Ye Wang, Kaihua Qin, Roger Wattenhofer, Dawn Song, and Arthur Gervais. Sok: Decentralized finance (defi) attacks. In 2023 IEEE Symposium on Security and Privacy (SP), pages 2444–

    2023

  72. [2461]

    Table 5: The attack incidents that are in the released datasets of FlashSyn and DeFiRanger but are not selected

    IEEE, 2023. Table 5: The attack incidents that are in the released datasets of FlashSyn and DeFiRanger but are not selected. Dataset Chain App Reason FlashSyn ETH Eminence Design Flaw ETH Yearn Design Flaw ETH bearFi Design Flaw BSC AutoShark Non-Contract BSC ElevenFi Design F...

    2023

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.