Pith. sign in

REVIEW 5 major objections 7 minor 75 references

Towards Understanding Deep Learning Model in Image Recognition via Coverage Test

T0 review · 5 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper claims that the six standard neuron-coverage metrics respond to model depth and coverage-configuration parameters in regular, describable ways, so a network's coverage values could be anticipated from its architecture and test…

desk verdict The empirical sweep is useful, but the paper's headline claim that coverage decreases with depth is contradicted by its own LeNet table. read the letter →

arxiv 2505.08814 v3 pith:TFW6PA4T submitted 2025-05-12 cs.CV cs.AIcs.LG

Wenkai Li , Xiaoqi Li , Yingjie Mao , Yishun Wang This is my paper · ORCID
classification cs.CVcs.AIcs.LG
keywords neuroncoverageDNNsecuritytestingmetricsmodeldepthMC/DCKMNCempiricalstudyimageclassifiers
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

Deep neural networks are tested by feeding them inputs and measuring how many neurons those inputs activate, but no established body of evidence says how those coverage numbers change as a network gets deeper or as test parameters change. This paper tries to fill that gap. It trains ten image classifiers, spanning LeNet, VGG, and ResNet variants from five to 54 layers, and measures six coverage metrics under varying thresholds, segmentation counts, boundary distances, and dataset sizes. The central claim is that coverage values follow regular patterns as model depth and configuration parameters move, with the headline rule being "as the model depth increases, the neuron coverage decreases." If the rules hold, security testers could predict coverage behavior from architecture choices instead of discovering it by running each model.

What carries the argument

The machinery is a controlled comparison of six coverage metrics across three architecture families. The metrics are the measuring instruments: neuron coverage (NC) counts neurons whose activation exceeds a threshold; k-multisection neuron coverage (KMNC) divides each neuron's activation range into k bins and counts occupied bins; neuron boundary coverage (NBC) and strong neuron activation coverage (SNAC) count activations outside the normal range, with SNAC restricted to the upper side; top-k neuron coverage (TopkNC) checks whether the most active neurons in each layer fall in the top k; and MC/DC treats a neuron as a decision and the previous layer's neurons as conditions. The comparison holds the dataset fixed within each architecture family, MNIST for LeNet and CIFAR10 for VGG and ResNet, and varies only depth, the metric's threshold or segmentation parameter, and, for MC/DC, the size of the input test set. The observed trends across these conditions are the argument's output.

What would settle it

Train the same ten architectures several times with different random seeds but identical data and training settings, then compute NC at thresholds from 0.3 to 0.9 for every checkpoint; if the depth ordering of coverage values changes across seeds, for instance VGG13 sometimes covering more than VGG11 at threshold 0.75, the claimed depth rules are not stable.

Watch

Extended reading notes

Core claim

On the authors' own terms, the discovery is an empirical regularity: the amount of activation space a test suite covers is not arbitrary but tracks model depth and metric configuration. In their data, neuron coverage (the fraction of neurons whose activation exceeds a threshold) falls as networks get deeper for the VGG and ResNet families at higher thresholds, while the small LeNet family instead shows coverage rising with depth; across all models, neuron coverage declines as the threshold rises and approaches a stable floor. K-multisection neuron coverage falls as the segmentation count k grows, with the drop steepest between k=10 and k=5000. Boundary coverage responds to the boundary-distance parameter in two regimes, depth-sensitive for negative distances and nearly flat for positive distances. Top-k neuron coverage is weakly tied to depth in VGG, negatively tied in ResNet, and in LeNet the coverage gap between successive depths shrinks as depth grows. Traditional MC/DC coverage grows with dataset size, with the value-value variant already saturated, and both MC/DC(SV) and MC/DC(VV) fall with model depth. The paper takes these patterns as evidence that depth and configuration information can be used to anticipate coverage behavior in DNN security testing.

Load-bearing premise

The experiments assume that one trained checkpoint per architecture and depth represents that configuration, so the depth trends are properties of depth rather than artifacts of a particular random initialization or training run.

Editorial extensions

If this is right

  • If the depth rules hold, coverage values for a given architecture are predictable from the layer count and the chosen threshold, so a tester can pick settings that expose under-covered regions before spending compute on inference.
  • The finding that MC/DC coverage rises with dataset size implies that small test suites systematically underreport structural coverage, which would bias comparisons between models tested on different data volumes.
  • The two-regime boundary behavior, depth-sensitive for negative boundary distance and nearly flat for positive distance, gives testers a parameter for targeting boundary neurons in either shallow or deep models.
  • Because KMNC's decline steepens between k=10 and k=5000, the segmentation count acts as a sensitivity knob that should be fixed when comparing models or reported alongside any coverage number.
  • The difference between sequential VGG and residual ResNet in top-k coverage suggests architecture family must be separated from depth in any coverage comparison.

Reading between the lines

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

  • If these depth rules generalize across random seeds, coverage could be modeled as a function of depth and threshold, letting researchers estimate coverage for architectures they have not yet trained.
  • The LeNet-on-MNIST and VGG/ResNet-on-CIFAR10 split leaves open a cross-dataset test: varying depth within a single dataset and architecture family would isolate depth from data difficulty.
  • A direct implication the paper does not draw is that if coverage systematically falls with depth, deeper models may appear less tested under fixed test suites, which could change how coverage is used as a quality gate in adversarial-robustness work.
  • The monotone patterns suggest coverage could serve as an architectural prior for test-input generation, for instance selecting per-layer thresholds to equalize coverage across depths.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 7 minor

Summary. The paper reports an empirical study of six coverage metrics (NC, KMNC, NBC, SNAC, Top-k NC, and MC/DC) across LeNet, VGG, and ResNet models of varying depths, using MNIST for LeNet and CIFAR-10 for VGG and ResNet. It claims to identify regular relationships between model depth, configuration parameters (activation threshold, partition count k, boundary distance, Top-k value, dataset size), and coverage, and it answers three research questions about these relationships. The central answer to RQ1 is that neuron coverage decreases as model depth increases, with additional statements about depth-dependent decline rates. The paper closes with three proposed future directions for DNN security testing.

Significance. The question of how architecture depth and configuration parameters affect coverage metrics is relevant to DNN testing, and the paper's tables contain direct measurements that could in principle be reused by other researchers. However, the significance of the reported findings is undercut by two problems: the headline depth rule is contradicted by the paper's own Table 2 for LeNet, and the experiments rely on a single trained checkpoint per configuration with no training details, no accuracies, and no repeated runs. As submitted, the empirical claims are not established, and the paper includes no code or data artifacts to support independent verification.

major comments (5)
  1. [Section 3.2, Answer to RQ1] The claim that "as the model depth increases, the neuron coverage decreases" is directly contradicted by Table 2 for LeNet. At every reported threshold the NC values strictly increase with depth: at tau=0.3 they are 0.5476 (5 layers), 0.7391 (6 layers), and 0.8256 (7 layers), and the same increasing ordering holds at tau=0.45, 0.6, 0.75, and 0.9. The text in Section 3.2 even states that "LeNet models with greater depth will have higher neuron coverage." The RQ1 answer therefore needs an explicit scope, such as "for VGG at thresholds above 0.6," or a reconciliation with the LeNet rows; as stated, the central conclusion is false on the paper's own data.
  2. [Section 3.1, Experimental Setup] The experiments are not reproducible from the manuscript: for each of the 10 models, a single trained checkpoint is used, and no training hyperparameters, optimizer, learning rate, epochs, data augmentation, or final test accuracy are reported. Tables 2-5 list one value per condition with no repeated seeds, error bars, or significance tests. Since coverage values can vary with initialization and training protocol, the depth trends derived from these tables could be artifacts of individual checkpoints; the paper should report accuracies and at least several independent training runs per configuration.
  3. [Section 3.2, Table 2] The comparison across architectures confounds model depth with dataset and architecture family: LeNet is evaluated on MNIST, while VGG and ResNet are evaluated on CIFAR-10, and the three families differ in layer types and connectivity. Consequently, statements such as "compared with the VGG model, the neuron coverage of ResNet decreases more obviously" do not isolate the effect of depth; depth, dataset, and architecture are varied simultaneously, so the reported cross-architecture depth rules are not identifiable.
  4. [Section 3.3, KMNC] The claim that "when the model is LeNet, the coverage of k-multisection neurons also decreases with the increase of the model depth" is not supported by Table 2. For k=10, KMNC goes 0.95 (5 layers), 0.8529 (6 layers), then 0.907 (7 layers); at k=1000 it goes 0.6023, 0.5433, and 0.591, and at k=5000 it goes 0.4464, 0.3493, and 0.3559. The non-monotonic pattern at several k values means the stated depth rule needs qualification or a statistical test rather than a blanket statement.
  5. [Section 3.7, Answer to RQ3] The claim that "all the coverage rates increase with the increase of the dataset size" overstates Table 5. For LeNet with 7 layers, MC/DC(VS) decreases from 0.1572 at 3200 samples to 0.1531 at 6400 samples, and MC/DC(VV) is essentially flat at 1.0 for the 5-layer model across all dataset sizes. The monotonicity claim should be restricted to the cases the table actually supports, or supported with a statistical test.
minor comments (7)
  1. [Section 3.6] The text says "Based on the data in Table 5," but the presented Top-k coverage values appear in Table 4, while Table 5 contains MC/DC; the cross-reference should be corrected.
  2. [Section 2] Several formulas in Section 2 are garbled by encoding issues (for example, the NC expression contains misplaced braces and an extra character), so the definitions are hard to parse; they should be typeset cleanly.
  3. [References] Many references are unrelated to the paper's topic (for example, [4], [6], [19], [40], [54], and several others concern smart contracts, NFTs, and blockchain), which suggests the bibliography was not curated for this submission; these should be removed or replaced with relevant DNN testing literature.
  4. [Figures] Figures 5, 6, 8, and 16 are referenced in Sections 3.2-3.6 but do not appear in the provided manuscript text, which currently contains only Figures 1-4; the figure numbering and placement must be fixed.
  5. [Abstract and Contributions] The abstract and the contributions list four coverage metrics, while the paper actually studies six (adding SNAC and Top-k NC); the counts should be made consistent.
  6. [Throughout] There are numerous typos and malformed glyph sequences, including "MD/DC" in Section 1, "neurons boundary scope of coverage concern" in Section 3.4, and the "/uni000000..." artifact strings in the captions of Figures 2-4; a careful editing pass is needed.
  7. [Reproducibility] The paper does not state whether code, trained models, or raw coverage measurements will be released; for an empirical study of this kind, a data and code availability statement is essential.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: all coverage values are directly measured and reported; no fitted parameter is renamed as a prediction.

full rationale

The paper is an empirical measurement study. Every coverage number in Tables 2-5 is obtained by running the stated coverage definitions on trained models and test inputs, then directly reported; there is no derivation chain in which an output quantity is defined in terms of the quantity it claims to explain. The LeNet 'stable value delta' statements in Section 3.2 are qualitative summaries of the same measured curves, not independent predictions, so they are descriptive rather than circular. The many self-citations in the reference list concern prior smart-contract and blockchain work and do not supply any load-bearing premise for the coverage experiments; the coverage metric definitions are attributed to external prior work such as DeepXplore, DeepGauge, and DeepConcolic. No uniqueness theorem is imported, and no ansatz is smuggled in via citation. The apparent contradiction between the RQ1 summary answer ('coverage decreases with depth') and the LeNet rows of Table 2, where NC increases with depth, is a correctness and generalization issue about the empirical claim, not a circularity issue, and the single-checkpoint, no-seed protocol is an experimental rigor concern. Neither affects the circularity score.

Assumptions & free parameters 6 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new physical or model entities. Its central claim rests on hand-chosen coverage configuration parameters, on the meaningfulness of published coverage metrics, and on the representativeness of single unseeded training runs across different datasets and architectures.

free parameters (6)
  • Activation threshold tau (NC) = 0.3, 0.45, 0.6, 0.75, 0.9
    Hand-chosen threshold in the neuron coverage definition; the reported depth trends depend on its range, for example LeNet coverage increases with depth at all listed tau while VGG decreases only for tau above 0.6.
  • KMNC partition count k = 10, 100, 1000, 5000, 10000
    Hand-chosen number of activation intervals; Table 2 shows coverage decreases as k increases.
  • Boundary distance epsilon = multiples of tau from -1.5*tau to 2*tau
    Hand-chosen offset for NBC and SNAC boundary regions; the depth trends differ for epsilon below zero and above zero.
  • Top-k rank threshold k = 5, 10, 15, 20, 25, 30, 35
    Hand-chosen number of top neurons per layer in Top-K neuron coverage.
  • MC/DC dataset size = 100, 200, 400, 800, 1600, 3200, 6400
    Hand-chosen test-set sizes to study MC/DC growth; these are experimental inputs, not fitted from data.
  • Stable coverage value delta for LeNet = about 0.24 for LeNet1, 0.58 for LeNet4, 0.64 for LeNet5
    The paper reads asymptotic coverage values off its own curves and uses them to state depth rules; these values are fitted to the same data they describe.
assumptions (3)
  • domain assumption Coverage metrics as defined in DeepGauge, DeepXplore, DeepHunter, and MC/DC are meaningful and correctly implemented proxies for DNN test adequacy.
    Section 2 defines metrics by citing prior tools; the paper does not independently validate the metrics or their implementations.
  • domain assumption A single trained LeNet, VGG, or ResNet checkpoint per depth is representative of that architecture and depth.
    Section 3.1 trains one model per configuration and does not report accuracy, hyperparameters, seeds, or repeated runs, yet the inferred depth rules treat each checkpoint as the architecture's behavior.
  • domain assumption Comparing LeNet trained on MNIST with VGG and ResNet trained on CIFAR10 isolates model-depth effects.
    Section 3.1 assigns different datasets to different architectures, confounding architecture, dataset, and depth effects; the RQ1 conclusions nevertheless compare across these groups.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Understanding Deep Learning Model in Image Recognition via Coverage Test." pith.science (2026). https://pith.science/paper/TFW6PA4T

@misc{pith2026250508814,
  author       = {Pith},
  title        = {Pith review of: Towards Understanding Deep Learning Model in Image Recognition via Coverage Test},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TFW6PA4T}},
  note         = {Machine review of arXiv:2505.08814}
}
read the original abstract

Deep neural networks (DNNs) play a crucial role in the field of artificial intelligence, and their security-related testing has been a prominent research focus. By inputting test cases, the behavior of models is examined for anomalies, and coverage metrics are utilized to determine the extent of neurons covered by these test cases. With the widespread application and advancement of DNNs, different types of neural behaviors have garnered attention, leading to the emergence of various coverage metrics for neural networks. However, there is currently a lack of empirical research on these coverage metrics, specifically in analyzing the relationships and patterns between model depth, configuration information, and neural network coverage. This paper aims to investigate the relationships and patterns of four coverage metrics: primary functionality, boundary, hierarchy, and structural coverage. A series of empirical experiments were conducted, selecting LeNet, VGG, and ResNet as different DNN architectures, along with 10 models of varying depths ranging from 5 to 54 layers, to compare and study the relationships between different depths, configuration information, and various neural network coverage metrics. Additionally, an investigation was carried out on the relationships between modified decision/condition coverage and dataset size. Finally, three potential future directions are proposed to further contribute to the security testing of DNN Models.

Figures

Figures reproduced from arXiv: 2505.08814 by the authors.

Figure 1
Figure 1. Comparison of traditional code and deep neural [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The NC and KMNC in Different Models. Blue groups [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The NBC and SNAC in Different Models. Blue groups [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: The TopKNC in Different Models. Blue groups rep [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The MC/DC Coverage. coverage in neurons are absolute value, then we get 𝜇(𝐿(5), 𝐿(6))− 𝜇(𝐿(6), 𝐿(7)) > 0. 𝐿(𝛾) represents the LeNet network structure with 𝛾 layers. Therefore, there is an important relationship between Top-k neuron coverage and model depth in LeNet, a …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

75 extracted references · 55 canonical work pages

  1. [1]

    Black-box testing of deep neural networks through test case diversity 2023.IEEE Transactions on Software Engineering49, 5, 3182–3204

    Zohreh Aghababaeyan, Manel Abdellatif, Lionel Briand, Mojtaba Bagherzadeh, and et al. Black-box testing of deep neural networks through test case diversity 2023.IEEE Transactions on Software Engineering49, 5, 3182–3204

    work page 2023

  2. [2]

    A systematic review on code clone detection 2019.IEEE access7, 86121–86144

    Qurat Ul Ain, Wasi Haider Butt, Muhammad Waseem Anwar, Farooque Azam, and Bilal Maqbool. A systematic review on code clone detection 2019.IEEE access7, 86121–86144

    work page 2019

  3. [3]

    The non-fungible token (NFT) market and its relationship with Bitcoin and Ethereum 2022.FinTech1, 3, 216–224

    Lennart Ante. The non-fungible token (NFT) market and its relationship with Bitcoin and Ethereum 2022.FinTech1, 3, 216–224

    work page 2022

  4. [4]

    Non-fungible token (NFT) markets on the Ethereum blockchain: Temporal development, cointegration and interrelations 2023.Economics of Innovation and New Technology32, 8, 1216–1234

    Lennart Ante. Non-fungible token (NFT) markets on the Ethereum blockchain: Temporal development, cointegration and interrelations 2023.Economics of Innovation and New Technology32, 8, 1216–1234

    work page 2023

  5. [5]

    Longformer: The long- document transformer 2020.arXiv preprint arXiv:2004.05150

    Iz Beltagy, Matthew E Peters, and Arman Cohan. Longformer: The long- document transformer 2020.arXiv preprint arXiv:2004.05150

    arXiv 2020

  6. [6]

    Formal verification of smart contracts: Short paper 2016

    Karthikeyan Bhargavan, Antoine Delignat-Lavaud, Cédric Fournet, Anitha Gol- lamudi, Georges Gonthier, Nadim Kobeissi, Natalia Kulatova, Aseem Rastogi, Thomas Sibut-Pinote, Nikhil Swamy, et al. Formal verification of smart contracts: Short paper 2016. InProceedings of the 2016 ACM workshop on programming languages and analysis for security. 91–96

    work page 2016

  7. [7]

    Enriching word vectors with subword information 2017.Transactions of the association for computational linguistics5, 135–146

    Piotr Bojanowski, Edouard Grave, Armand Joulin, and Tomas Mikolov. Enriching word vectors with subword information 2017.Transactions of the association for computational linguistics5, 135–146

    work page 2017

  8. [8]

    Enhancing smart contract vulnerability detection in dapps leveraging fine-tuned llm 2025.arXiv preprint arXiv:2504.05006

    Jiuyang Bu, Wenkai Li, Zongwei Li, Zeng Zhang, and Xiaoqi Li. Enhancing smart contract vulnerability detection in dapps leveraging fine-tuned llm 2025.arXiv preprint arXiv:2504.05006

    arXiv 2025

Show all 75 references
  1. [9]

    SmartBugBert: BERT-Enhanced Vulnerability Detection for Smart Contract Bytecode 2025.arXiv preprint arXiv:2504.05002

    Jiuyang Bu, Wenkai Li, Zongwei Li, Zeng Zhang, and Xiaoqi Li. SmartBugBert: BERT-Enhanced Vulnerability Detection for Smart Contract Bytecode 2025.arXiv preprint arXiv:2504.05002

    2025

  2. [10]

    Deepinspect: A black-box trojan detection and mitigation framework for deep neural networks

    Huili Chen, Cheng Fu, Jishen Zhao, and Farinaz Koushanfar. Deepinspect: A black-box trojan detection and mitigation framework for deep neural networks

  3. [11]

    Rethinking attention with performers 2020.arXiv preprint arXiv:2009.14794

    Krzysztof Choromanski, Valerii Likhosherstov, David Dohan, Xingyou Song, Andreea Gane, Tamas Sarlos, Peter Hawkins, Jared Davis, Afroz Mohiuddin, Lukasz Kaiser, et al. Rethinking attention with performers 2020.arXiv preprint arXiv:2009.14794

    2020 arXiv

  4. [12]

    A survey on smart contract vulnerabilities: Data sources, detection and repair 2023.Information and Software Technology159, 107221

    Hanting Chu, Pengcheng Zhang, Hai Dong, Yan Xiao, Shunhui Ji, and Wenrui Li. A survey on smart contract vulnerabilities: Data sources, detection and repair 2023.Information and Software Technology159, 107221

    2023

  5. [13]

    SmartBugs: A Framework to Analyze Solidity Smart Contracts 2020

    João F Ferreira, Pedro Cruz, Thomas Durieux, and Rui Abreu. SmartBugs: A Framework to Analyze Solidity Smart Contracts 2020. InProceedings of the 35th 8 Towards Understanding Deep Learning Model in Image Recognition via Coverage Test Conference’17, July 2017, Washington, DC, U...

    2020

  6. [14]

    Checking Smart Contracts with Structural Code Embedding 2020.IEEE Transactions on Software Engineering

    Zhipeng Gao, Lingxiao Jiang, Xin Xia, David Lo, and John Grundy. Checking Smart Contracts with Structural Code Embedding 2020.IEEE Transactions on Software Engineering

    2020

  7. [15]

    How effective are smart contract analysis tools? evaluating smart contract static analysis tools using bug injection

    Asem Ghaleb and Karthik Pattabiraman. How effective are smart contract analysis tools? evaluating smart contract static analysis tools using bug injection

  8. [16]

    Achecker: Statically de- tecting smart contract access control vulnerabilities 2023

    Asem Ghaleb, Julia Rubin, and Karthik Pattabiraman. Achecker: Statically de- tecting smart contract access control vulnerabilities 2023. In2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE). IEEE, 945–956

    2023

  9. [17]

    Explaining and harnessing adversarial examples 2014.arXiv preprint arXiv:1412.6572

    Ian J Goodfellow, Jonathon Shlens, and Christian Szegedy. Explaining and harnessing adversarial examples 2014.arXiv preprint arXiv:1412.6572

    2014 arXiv

  10. [18]

    Deep residual learning for image recognition 2016

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition 2016. InProceedings of the IEEE conference on computer vision and pattern recognition (CVPR). 770–778

    2016

  11. [19]

    Characterizing code clones in the ethereum smart contract ecosystem 2020

    Ningyu He, Lei Wu, Haoyu Wang, Yao Guo, and Xuxian Jiang. Characterizing code clones in the ethereum smart contract ecosystem 2020. InFinancial Cryp- tography and Data Security: 24th International Conference, FC 2020, Kota Kinabalu, Malaysia, February 10–14, 2020 Revised Selec...

    2020

  12. [20]

    Hunting vulnerable smart contracts via graph embedding based bytecode matching 2021.IEEE Transactions on Information Forensics and Security16, 2144–2156

    Jianjun Huang, Songming Han, Wei You, Wenchang Shi, Bin Liang, Jingzheng Wu, and Yanjun Wu. Hunting vulnerable smart contracts via graph embedding based bytecode matching 2021.IEEE Transactions on Information Forensics and Security16, 2144–2156

    2021

  13. [21]

    Characterizing the Solana NFT ecosys- tem 2024

    Dechao Kong, Xiaoqi Li, and Wenkai Li. Characterizing the Solana NFT ecosys- tem 2024. InCompanion Proceedings of the ACM Web Conference 2024. 766–769

    2024

  14. [22]

    Neural network models and deep learning 2019.Current Biology29, 7, R231–R236

    Nikolaus Kriegeskorte and Tal Golan. Neural network models and deep learning 2019.Current Biology29, 7, R231–R236

    2019

  15. [23]

    Gradient-based learning applied to document recognition 1998.Proc

    Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition 1998.Proc. IEEE86, 11, 2278–2324

    1998

  16. [24]

    Cobra: interaction-aware bytecode-level vulnerability detector for smart contracts 2024

    Wenkai Li, Xiaoqi Li, Zongwei Li, and Yuqing Zhang. Cobra: interaction-aware bytecode-level vulnerability detector for smart contracts 2024. InProceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering. 1358–1369

    2024

  17. [25]

    Detecting Malicious Accounts in Web3 through Transaction Graph 2024

    Wenkai Li, Zhijie Liu, Xiaoqi Li, and Sen Nie. Detecting Malicious Accounts in Web3 through Transaction Graph 2024. InProceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering. 2482–2483

    2024

  18. [26]

    Hybrid analysis of smart contracts and malicious behaviors in ethereum 2021

    Xiaoqi Li et al. Hybrid analysis of smart contracts and malicious behaviors in ethereum 2021

    2021

  19. [27]

    CLUE: towards discovering locked cryptocurrencies in ethereum 2021

    Xiaoqi Li, Ting Chen, Xiapu Luo, and Chenxu Wang. CLUE: towards discovering locked cryptocurrencies in ethereum 2021. InProceedings of the 36th Annual ACM Symposium on Applied Computing. 1584–1587

    2021

  20. [28]

    Xiaoqi Li, L Yu, and XP Luo. 2017. On Discovering Vulnerabilities in Android Applications. InMobile Security and Privacy. Elsevier, 155–166

    2017

  21. [29]

    ModelDiff: Testing-based DNN similarity comparison for model reuse detection 2021

    Yuanchun Li, Ziqi Zhang, Bingyan Liu, Ziyue Yang, and Yunxin Liu. ModelDiff: Testing-based DNN similarity comparison for model reuse detection 2021. In Proceedings of the 30th ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA). 139–151

    2021

  22. [30]

    StateGuard: Detecting State Derailment Defects in Decentralized Exchange Smart Contract 2024

    Zongwei Li, Wenkai Li, Xiaoqi Li, and Yuqing Zhang. StateGuard: Detecting State Derailment Defects in Decentralized Exchange Smart Contract 2024. In Companion Proceedings of the ACM Web Conference 2024. 810–813

    2024

  23. [31]

    SCALM: Detecting Bad Practices in Smart Contracts Through LLMs 2025.arXiv preprint arXiv:2502.04347

    Zongwei Li, Xiaoqi Li, Wenkai Li, and Xin Wang. SCALM: Detecting Bad Practices in Smart Contracts Through LLMs 2025.arXiv preprint arXiv:2502.04347

    2025

  24. [32]

    On identity, transaction, and smart contract privacy on permissioned and per- missionless blockchain: A comprehensive survey 2024.Comput

    Wei Liang, Yaqin Liu, Ce Yang, Songyou Xie, Kuanching Li, and Willy Susilo. On identity, transaction, and smart contract privacy on permissioned and per- missionless blockchain: A comprehensive survey 2024.Comput. Surveys56, 12, 1–35

    2024

  25. [33]

    SoK: Security Analysis of Blockchain-based Cryptocur- rency 2025.arXiv preprint arXiv:2503.22156

    Zekai Liu and Xiaoqi Li. SoK: Security Analysis of Blockchain-based Cryptocur- rency 2025.arXiv preprint arXiv:2503.22156

    2025 arXiv

  26. [34]

    GasTrace: Detecting Sand- wich Attack Malicious Accounts in Ethereum 2024

    Zekai Liu, Xiaoqi Li, Hongli Peng, and Wenkai Li. GasTrace: Detecting Sand- wich Attack Malicious Accounts in Ethereum 2024. In2024 IEEE International Conference on Web Services (ICWS). IEEE, 1409–1411

    2024

  27. [35]

    Deepgauge: Multi-granularity testing criteria for deep learning systems 2018

    Lei Ma, Felix Juefei-Xu, Fuyuan Zhang, Jiyuan Sun, Minhui Xue, Bo Li, Chunyang Chen, Ting Su, Li Li, Yang Liu, and et al. Deepgauge: Multi-granularity testing criteria for deep learning systems 2018. InProceedings of the 33rd ACM/IEEE international conference on automated soft...

    2018

  28. [36]

    Combining Fine-Tuning and LLM-based Agents for Intuitive Smart Contract Auditing with Justifications 2024.arXiv preprint arXiv:2403.16073

    Wei Ma, Daoyuan Wu, Yuqiang Sun, Tianwen Wang, Shangqing Liu, Jian Zhang, Yue Xue, and Yang Liu. Combining Fine-Tuning and LLM-based Agents for Intuitive Smart Contract Auditing with Justifications 2024.arXiv preprint arXiv:2403.16073

    2024 arXiv

  29. [37]

    Yu A Malkov and Dmitry A Yashunin. Efficient and robust approximate nearest neighbor search using hierarchical navigable small world graphs 2018.IEEE transactions on pattern analysis and machine intelligence42, 4, 824–836

    2018

  30. [38]

    SCLA: Automated Smart Contract Summarization via LLMs and Semantic Augmentation 2024

    Yingjie Mao, Xiaoqi Li, Wenkai Li, Xin Wang, and Lei Xie. SCLA: Automated Smart Contract Summarization via LLMs and Semantic Augmentation 2024. arXiv preprint arXiv:2402.04863

    2024 arXiv

  31. [39]

    Efficient estimation of word representations in vector space 2013.arXiv preprint arXiv:1301.3781

    Tomas Mikolov. Efficient estimation of word representations in vector space 2013.arXiv preprint arXiv:1301.3781

    2013 arXiv

  32. [40]

    Mapping the NFT revolution: market trends, trade networks, and visual features 2021.Scientific reports11, 1, 20902

    Matthieu Nadini, Laura Alessandretti, Flavio Di Giacinto, Mauro Martino, Luca Maria Aiello, and Andrea Baronchelli. Mapping the NFT revolution: market trends, trade networks, and visual features 2021.Scientific reports11, 1, 20902

    2021

  33. [41]

    Understanding source code evolution using abstract syntax tree matching 2005

    Iulian Neamtiu, Jeffrey S Foster, and Michael Hicks. Understanding source code evolution using abstract syntax tree matching 2005. InProceedings of the 2005 international workshop on Mining software repositories. 1–5

    2005

  34. [42]

    Do wide and deep networks learn the same things? uncovering how neural network representations vary with width and depth 2020.arXiv preprint arXiv:2010.15327

    Thao Nguyen, Maithra Raghu, and Simon Kornblith. Do wide and deep networks learn the same things? uncovering how neural network representations vary with width and depth 2020.arXiv preprint arXiv:2010.15327

    2020 arXiv

  35. [43]

    Unveiling wash trading in popular NFT markets 2024

    Yuanzheng Niu, Xiaoqi Li, Hongli Peng, and Wenkai Li. Unveiling wash trading in popular NFT markets 2024. InCompanion Proceedings of the ACM Web Conference

    2024

  36. [44]

    Enhancing Ethereum smart-contracts static analysis by computing a precise Control-Flow Graph of Ethereum bytecode 2023.Journal of Systems and Software200, 111653

    Michele Pasqua, Andrea Benini, Filippo Contro, Marco Crosara, Mila Dalla Preda, and Mariano Ceccato. Enhancing Ethereum smart-contracts static analysis by computing a precise Control-Flow Graph of Ethereum bytecode 2023.Journal of Systems and Software200, 111653

    2023

  37. [45]

    Deepxplore: Automated whitebox testing of deep learning systems 2017

    Kexin Pei, Yinzhi Cao, Junfeng Yang, and Suman Jana. Deepxplore: Automated whitebox testing of deep learning systems 2017. Inproceedings of the 26th Sym- posium on Operating Systems Principles (SOSP). 1–18

    2017

  38. [46]

    Smart contract vulnerability detection technique: A survey 2022

    Peng Qian, Zhenguang Liu, Qinming He, Butian Huang, Duanzheng Tian, and Xun Wang. Smart contract vulnerability detection technique: A survey 2022. arXiv preprint arXiv:2209.05872

    2022 arXiv

  39. [47]

    Sourcerercc: Scaling code clone detection to big-code 2016

    Hitesh Sajnani, Vaibhav Saini, Jeffrey Svajlenko, Chanchal K Roy, and Cristina V Lopes. Sourcerercc: Scaling code clone detection to big-code 2016. InProceedings of the 38th international conference on software engineering. 1157–1168

    2016

  40. [48]

    An empirical study on test case prioritization metrics for deep neural networks 2021

    Ying Shi, Beibei Yin, Zheng Zheng, and Tiancheng Li. An empirical study on test case prioritization metrics for deep neural networks 2021. InProceedings of the IEEE 21st International Conference on Software Quality, Reliability and Security (QRS). 157–166

    2021

  41. [49]

    Very deep convolutional networks for large-scale image recognition 2014.arXiv preprint arXiv:1409.1556

    Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition 2014.arXiv preprint arXiv:1409.1556

    2014 arXiv

  42. [50]

    Fast graph attention networks using effective resistance based graph sparsification 2020.arXiv preprint arXiv:2006.08796

    Rakshith S Srinivasa, Cao Xiao, Lucas Glass, Justin Romberg, and Jimeng Sun. Fast graph attention networks using effective resistance based graph sparsification 2020.arXiv preprint arXiv:2006.08796

    2020 arXiv

  43. [51]

    Testing deep neural networks 2018.arXiv preprint arXiv:1803.04792

    Youcheng Sun, Xiaowei Huang, Daniel Kroening, James Sharp, Matthew Hill, and Rob Ashmore. Testing deep neural networks 2018.arXiv preprint arXiv:1803.04792

    2018 arXiv

  44. [52]

    DeepConcolic: Testing and debugging deep neural networks

    Youcheng Sun, Xiaowei Huang, Daniel Kroening, James Sharp, Matthew Hill, and Rob Ashmore. DeepConcolic: Testing and debugging deep neural networks

  45. [53]

    Structural test coverage criteria for deep neural networks 2019.ACM Transactions on Embedded Computing Systems (TECS)18, 5s, 1–23

    Youcheng Sun, Xiaowei Huang, Daniel Kroening, James Sharp, Matthew Hill, and Rob Ashmore. Structural test coverage criteria for deep neural networks 2019.ACM Transactions on Embedded Computing Systems (TECS)18, 5s, 1–23

    2019

  46. [54]

    Smart contracts: building blocks for digital markets 1996.EXTROPY: The Journal of Transhumanist Thought,(16)18, 2, 28

    Nick Szabo. Smart contracts: building blocks for digital markets 1996.EXTROPY: The Journal of Transhumanist Thought,(16)18, 2, 28

    1996

  47. [55]

    InProceedings of the IEEE/ACM 41st International Conference on Software Engineering (ICSE). 111–114

  48. [56]

    Smartcheck: Static analysis of ethereum smart contracts 2018

    Sergei Tikhomirov, Ekaterina Voskresenskaya, Ivan Ivanitskiy, Ramil Takhaviev, Evgeny Marchenko, and Yaroslav Alexandrov. Smartcheck: Static analysis of ethereum smart contracts 2018. InProceedings of the 1st international workshop on emerging trends in software engineering fo...

    2018

  49. [57]

    A survey of smart contract formal specification and verification 2021.ACM Computing Surveys (CSUR)54, 7, 1–38

    Palina Tolmach, Yi Li, Shang-Wei Lin, Yang Liu, and Zengxiang Li. A survey of smart contract formal specification and verification 2021.ACM Computing Surveys (CSUR)54, 7, 1–38

    2021

  50. [58]

    Ethereum Smart Contract Representation Learning for Robust Bytecode-Level Similarity Detection

    Zhenzhou Tian, Yaqian Huang, Jie Tian, Zhongmin Wang, Yanping Chen, and Lingwei Chen. Ethereum Smart Contract Representation Learning for Robust Bytecode-Level Similarity Detection. 2022. InSEKE. 513–518

    2022

  51. [59]

    Non-fungible token (NFT): Overview, evaluation, opportunities and challenges 2021.arXiv preprint arXiv:2105.07447

    Qin Wang, Rujia Li, Qi Wang, and Shiping Chen. Non-fungible token (NFT): Overview, evaluation, opportunities and challenges 2021.arXiv preprint arXiv:2105.07447

    2021 arXiv

  52. [60]

    Smart contracts in the real world: A statistical exploration of external data dependencies 2024.arXiv preprint arXiv:2406.13253

    Yishun Wang, Xiaoqi Li, Shipeng Ye, Lei Xie, and Ju Xing. Smart contracts in the real world: A statistical exploration of external data dependencies 2024.arXiv preprint arXiv:2406.13253

    2024

  53. [61]

    Securify: Practical security analysis of smart con- tracts 2018

    Petar Tsankov, Andrei Dan, Dana Drachsler-Cohen, Arthur Gervais, Florian Buenzli, and Martin Vechev. Securify: Practical security analysis of smart con- tracts 2018. InProceedings of the 2018 ACM SIGSAC conference on computer and communications security. 67–82

    2018

  54. [62]

    WakeMint: Detecting Sleep- minting Vulnerabilities in NFT Smart Contracts 2025

    Lei Xiao, Shuo Yang, Wen Chen, and Zibin Zheng. WakeMint: Detecting Sleep- minting Vulnerabilities in NFT Smart Contracts 2025. In2025 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER). IEEE, 740–750

    2025

  55. [63]

    Npc: Neuron path coverage via characterizing decision logic of deep neural 9 Conference’17, July 2017, Washington, DC, USA Wenkai and Xiaoqi, et al

    Xiaofei Xie, Tianlin Li, Jian Wang, Lei Ma, Qing Guo, Felix Juefei-Xu, and Yang Liu. Npc: Neuron path coverage via characterizing decision logic of deep neural 9 Conference’17, July 2017, Washington, DC, USA Wenkai and Xiaoqi, et al. networks 2022.ACM Transactions on Software ...

    2017

  56. [64]

    Deep learning code fragments for code clone detection 2016

    Martin White, Michele Tufano, Christopher Vendome, and Denys Poshyvanyk. Deep learning code fragments for code clone detection 2016. InProceedings of the 31st IEEE/ACM international conference on automated software engineering. 87–98

    2016

  57. [65]

    Correlations between deep neural network model coverage criteria and model quality 2020

    Shenao Yan, Guanhong Tao, Xuwei Liu, Juan Zhai, Shiqing Ma, Lei Xu, and Xiangyu Zhang. Correlations between deep neural network model coverage criteria and model quality 2020. InProceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium...

    2020

  58. [66]

    Un- cover the premeditated attacks: Detecting exploitable reentrancy vulnerabilities by identifying attacker contracts 2024

    Shuo Yang, Jiachi Chen, Mingyuan Huang, Zibin Zheng, and Yuan Huang. Un- cover the premeditated attacks: Detecting exploitable reentrancy vulnerabilities by identifying attacker contracts 2024. InProceedings of the IEEE/ACM 46th International Conference on Software Engineering. 1–12

    2024

  59. [67]

    Deephunter: a coverage-guided fuzz testing framework for deep neural networks 2019

    Xiaofei Xie, Lei Ma, Felix Juefei-Xu, Minhui Xue, Hongxu Chen, Yang Liu, Jianjun Zhao, Bo Li, Jianxiong Yin, and Simon See. Deephunter: a coverage-guided fuzz testing framework for deep neural networks 2019. InProceedings of the 28th ACM SIGSOFT international symposium on soft...

    2019

  60. [68]

    Combining GPT and Code-Based Similarity Checking for Effective Smart Contract Vulnerability Detection 2024.arXiv preprint arXiv:2412.18225

    Jango Zhang. Combining GPT and Code-Based Similarity Checking for Effective Smart Contract Vulnerability Detection 2024.arXiv preprint arXiv:2412.18225

    2024 arXiv

  61. [69]

    Acfix: Guiding llms with mined common rbac practices for context- aware repair of access control vulnerabilities in smart contracts 2024.arXiv preprint arXiv:2403.06838

    Lyuye Zhang, Kaixuan Li, Kairan Sun, Daoyuan Wu, Ye Liu, Haoye Tian, and Yang Liu. Acfix: Guiding llms with mined common rbac practices for context- aware repair of access control vulnerabilities in smart contracts 2024.arXiv preprint arXiv:2403.06838

    2024 arXiv

  62. [70]

    Revisiting neuron coverage for dnn testing: A layer-wise and distribution-aware criterion 2023

    Yuanyuan Yuan, Qi Pang, and Shuai Wang. Revisiting neuron coverage for dnn testing: A layer-wise and distribution-aware criterion 2023. InProceedings of the IEEE/ACM 45th International Conference on Software Engineering (ICSE). 1200–1212

    2023

  63. [71]

    Byte- code similarity detection of smart contract across optimization options and compiler versions based on triplet network 2022.Electronics11, 4, 597

    Di Zhu, Feng Yue, Jianmin Pang, Xin Zhou, Wenjie Han, and Fudong Liu. Byte- code similarity detection of smart contract across optimization options and compiler versions based on triplet network 2022.Electronics11, 4, 597

    2022

  64. [72]

    Malicious Code Detection in Smart Contracts via Opcode Vectorization 2025.arXiv preprint arXiv:2504.12720

    Huanhuan Zou, Zongwei Li, and Xiaoqi Li. Malicious Code Detection in Smart Contracts via Opcode Vectorization 2025.arXiv preprint arXiv:2504.12720. 10

    2025

  65. [73]

    PrettySmart: Detecting Permission Re-delegation Vulnerability for To- ken Behaviors in Smart Contracts 2024

    Zhijie Zhong, Zibin Zheng, Hong-Ning Dai, Qing Xue, Junjia Chen, and Yuhong Nan. PrettySmart: Detecting Permission Re-delegation Vulnerability for To- ken Behaviors in Smart Contracts 2024. InProceedings of the IEEE/ACM 46th International Conference on Software Engineering. 1–12

    2024

  66. [2019]

    InProceedings of the International Joint Conferences on Artificial Intelligence, Vol. 2. 8

  67. [2020]

    InProceedings of the 29th ACM SIGSOFT international symposium on software testing and analysis. 415–427

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.