Pith. sign in

REVIEW 4 major objections 6 minor 92 references

A Survey: Towards Privacy and Security in Mobile Large Language Models

T0 review · 4 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read This survey claims that the privacy and security problems of mobile large language models can be organized into four data-privacy defenses and four attack classes, and that mobile resource limits decide which defenses are practical.

desk verdict A well-organized but shallow survey that botches its side-channel defense section and overclaims comprehensiveness. read the letter →

arxiv 2509.02411 v1 pith:5TDUC7LS submitted 2025-09-02 cs.CR cs.AI

Honghui Xu , Kaiyang Li , Wei Chen , Danyang Zheng , Zhiyuan Li , Zhipeng Cai This is my paper
classification cs.CRcs.AI
keywords mobilelargelanguagemodelsLLMprivacysecuritydifferentialfederatedlearningpromptencryptionmembershipinferenceattacksadversarial
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 is a survey that tries to establish a usable map of privacy and security issues for large language models running on phones and other mobile devices. It claims that existing defenses fall into four families—data anonymization, prompt encryption, differential privacy, and federated learning—and that the main attacks are adversarial, membership inference, model inversion, and side-channel. It compares each defense's effectiveness and limitations, and argues that mobile constraints such as limited compute, battery, and memory make server-style protections expensive or impractical. The value of the survey, if correct, is a single structured reference for developers, policymakers, and researchers who need to know what threats exist and which mitigations can realistically run on-device.

What carries the argument

The load-bearing device is the paper's two-part taxonomy, introduced in Figure 1 and developed in Sections III through V. It separates data privacy preservation—anonymization, prompt encryption, differential privacy, and federated learning—from security countermeasures against adversarial, membership inference, model inversion, and side-channel attacks. The taxonomy works by framing each defense primarily through the mobile constraint: sensitive data is processed at the edge, where devices lack the compute, battery, and memory for heavyweight protections, so every method is evaluated on both its security effect and its feasibility on resource-limited hardware.

What would settle it

Find a well-known area of on-device model protection—such as trusted execution environments or secure hardware enclaves for mobile inference—and check whether the survey covers it; a substantial absent literature would contradict the comprehensiveness claim. A reader could also rerun a structured literature search with explicit queries and compare the resulting set with the survey's references.

Watch

Extended reading notes

Core claim

In the paper's own terms, the central claim is that privacy and security issues in mobile LLMs can be systematically categorized into a small set of solution families and attack classes. The solution families are data anonymization, prompt encryption, differential privacy, and federated learning; the attack classes are adversarial attacks, membership inference attacks, model inversion attacks, and side-channel attacks. The paper asserts that each defense has known limitations under mobile resource constraints—differential privacy can degrade model accuracy, encryption and local processing strain battery and memory, and federated learning introduces communication and heterogeneity challenges—

Load-bearing premise

The survey's claim of comprehensiveness rests on the assumption that the papers it chose to cite represent the whole field of mobile LLM privacy and security, since it does not document a search protocol or inclusion criteria.

Editorial extensions

If this is right

  • A developer choosing a privacy technique can use the survey's categories to see the main cost of each option: anonymization trades readability, encryption adds computation, differential privacy trades accuracy for noise, and federated learning spreads computation and data across devices.
  • Membership inference and model inversion are treated as serious risks for on-device LLMs because training data can include messages, voice commands, and location; the paper links defenses such as differential privacy, distillation, and adaptive regularization to countering overfitting.
  • The comparison implies that no single defense is sufficient, and that combinations—such as federated learning with differentially private parameter-efficient fine-tuning—are the current direction for balancing privacy and efficiency.
  • The open problems identified in the paper—mobile-friendly adversarial robustness, lightweight standardized privacy frameworks, and secure low-latency transmission—define a near-term research agenda for trustworthy mobile LLM deployment.

Reading between the lines

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

  • Beyond the paper: the map is likely illustrative rather than exhaustive. A documented search protocol with inclusion and exclusion criteria would be needed to confirm the claimed comprehensiveness; the paper does not provide one.
  • Beyond the paper: the taxonomy suggests a natural fifth defense family for hardware-assisted isolation, such as trusted execution environments or secure enclaves for on-device inference, which the survey does not cover.
  • Beyond the paper: the qualitative trade-off discussion could be turned into a concrete benchmark measuring latency, energy, memory, and accuracy loss of each defense on representative mobile hardware; the paper reports no such measurements.
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

4 major / 6 minor

Summary. The paper is a survey of privacy and security issues for large language models deployed on mobile and edge platforms. It reviews background material on LLM architecture and mobile deployment, classifies privacy-preserving techniques (data anonymization, prompt encryption, differential privacy, federated learning), describes defenses against four attack categories (adversarial, membership inference, model inversion, side-channel), and discusses applications in healthcare, finance, and education. The abstract claims both a comprehensive overview and an in-depth comparison of the effectiveness and limitations of these solutions.

Significance. If the survey were comprehensive and accurate, it would be a useful reference for practitioners and researchers working on mobile LLM security, a timely topic for an IoT-oriented venue. The paper does bring together a number of relevant references on federated learning and privacy-preserving fine-tuning, and it sketches application-specific trade-offs. However, the value of the survey depends on the reliability of its categorizations and the support for its stated contributions. As detailed below, several load-bearing statements are inaccurate or unsupported, and the claimed comprehensiveness and comparative depth are not delivered.

major comments (4)
  1. [V.D (Defense on Side-Channel Attack)] The section states that differential privacy is 'one of the most effective mitigations against the side-channel attacks' and cites [80]. Reference [80] is 'LLM-FIN: Large Language Models Fingerprinting Attack on Edge Devices', an attack paper, not a DP defense. Conceptually, ε-DP bounds the change in an output distribution given a change in one input; it does not constrain physical side-channel signals such as power, EM, timing, or memory access patterns observed during on-device inference. Federated learning, also listed, affects training-time data sharing and cannot prevent leakage during local inference. Since side-channel attacks are one of the four vulnerability classes named in the abstract, this is a load-bearing correctness error. Please replace these claims with supported defenses (e.g., masking, constant-time execution, ORAM, TEE-based isolation) or explicitly note the lack of
  2. [Abstract and overall methodology] The abstract claims a 'comprehensive overview' and 'systematically categorizing existing solutions.' No search protocol, inclusion/exclusion criteria, database list, time span, or quality assessment is provided. The selection of roughly 90 references appears ad hoc, and several relevant areas for mobile LLM privacy/security are absent, including trusted execution environments for on-device inference, secure inference protocols, and model extraction defenses. Please either add a methodology subsection describing the literature search and selection process, or substantially qualify the comprehensiveness claim.
  3. [Abstract and Sections IV-V] The abstract promises an 'in-depth comparison of their effectiveness and limitations' for the analyzed attacks and defenses. The body does not deliver this. Each defense receives a short qualitative paragraph, with no comparative metrics, no tables contrasting privacy guarantees, utility costs, computational overhead, or applicable attack scenarios, and no explicit mapping between threats and defenses. For example, differential privacy is presented as a defense against adversarial, membership inference, model inversion, and side-channel attacks, but the conditions under which it is effective (e.g., training-time DP-SGD vs. inference-time noise) are not compared or delimited. Please add a comparative analysis (e.g., a threat-defense matrix) or revise the abstract to reflect the actual scope.
  4. [VI (Trustworthy Mobile LLMs Practices)] Section VI opens by promising 'practical examples of successful LLM implementations that uphold data security and privacy.' The subsections that follow summarize general trade-offs and cite works that discuss privacy techniques; they do not document concrete deployed systems or evaluate their success. This overstatement is part of the paper's claimed contribution. Please either provide actual case studies with system details, or reframe the section as 'application areas and trade-offs' without the 'successful implementations' claim.
minor comments (6)
  1. [V.C / Fig. 6] Figure 6 is captioned 'Membership Inversion Attack' while the text refers to 'Model Inversion Attack'. The caption should be corrected.
  2. [II.B.2] Reference [38] is about an IoTDB data analysis library and appears unrelated to the claim that mobile devices transmit input data to edge servers to reduce resource strain. Please check and replace with a relevant citation.
  3. [IV.C] The sentence 'Differential Privacy applies controlled noise to datasets' is imprecise. Differential privacy is a formal definition; mechanisms such as DP-SGD add noise to gradients or outputs. Consider rewording to avoid conflating the definition with an algorithm.
  4. [V.D] Minor grammatical issue: 'can help mitigate to many security risks' should be 'can help mitigate many security risks.'
  5. [II.B.2 / Abstract] Section II.B.2 states the survey 'focuses primarily on examining the privacy and security challenges associated with the collaborative edge computing of LLMs,' but the abstract and later sections also emphasize on-device deployment. This scope statement should be clarified or aligned with the actual coverage.
  6. [VI.A] The text says 'methods like differential privacy protect data effectively, they can impair model performance' — the pronoun and sentence structure should be corrected for readability.

Circularity Check

0 steps flagged · score 2.0 of 10

No substantive circularity: the survey has no derivation chain; the few self-citations are background references and do not make any claim reduce to its own inputs.

full rationale

This is a survey paper, so the central claims are taxonomical and descriptive rather than derived. The abstract's promise of a 'comprehensive overview' and 'in-depth comparison' is a scope statement, not a mathematical or empirical derivation, and no equation or fitted parameter is involved. The self-citations that do appear are used as ordinary literature support in background and solution-survey contexts. For example, Section II.B says 'Recently, deploying LLMs on mobile computing platforms has become a trend ... [1], [2]', and Section IV.C says 'advanced strategies utilize selective privacy application during multiple fine-tuning phases of LLMs to enhance the balance between privacy and utility [8]'. These citations support existing background or survey claims rather than serving as the paper's own load-bearing derivation; the cited prior survey [8] is not used to prove the current survey's organizing claims, and no result is defined in terms of itself. The most notable defect is in Section V.D, where the claim 'One of the most effective mitigations against the side-channel attacks is differential privacy [80]' cites [80], an attack paper rather than a differential-privacy defense; the same paragraph cites the authors' [81] for a federated-learning mitigation. That is a correctness/support issue, not circularity: the paper does not define a concept in terms of a predicted quantity, nor does it relabel a fitted parameter as a prediction. The lack of a described systematic search protocol weakens the comprehensiveness claim but is also not a circular-reasoning problem. Because the survey contains no derivation chain and the self-citations are minor and non-load-bearing, the circularity score is low.

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

This is a survey, so there are no fitted parameters or invented entities. The central claim rests on background assumptions about LLM technology and on the accuracy of the primary literature being summarized.

assumptions (4)
  • domain assumption The transformer architecture is the dominant LLM architecture.
    The paper's background section assumes transformers without argument; this is a statement about current practice, not a mathematical theorem.
  • domain assumption Pre-training followed by fine-tuning is the standard training pipeline.
    Used in Section II-A3 to frame all later discussion of privacy-preserving fine-tuning.
  • domain assumption The cited papers accurately report their results.
    The survey is a synthesis; any misrepresentation in a cited source propagates.
  • domain assumption Security techniques from non-mobile settings are applicable to mobile constraints.
    The survey maps defenses like DP-SGD, LoRA, and FL to mobile devices without testing their resource feasibility.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Survey: Towards Privacy and Security in Mobile Large Language Models." pith.science (2026). https://pith.science/paper/5TDUC7LS

@misc{pith2026250902411,
  author       = {Pith},
  title        = {Pith review of: A Survey: Towards Privacy and Security in Mobile Large Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5TDUC7LS}},
  note         = {Machine review of arXiv:2509.02411}
}
read the original abstract

Mobile Large Language Models (LLMs) are revolutionizing diverse fields such as healthcare, finance, and education with their ability to perform advanced natural language processing tasks on-the-go. However, the deployment of these models in mobile and edge environments introduces significant challenges related to privacy and security due to their resource-intensive nature and the sensitivity of the data they process. This survey provides a comprehensive overview of privacy and security issues associated with mobile LLMs, systematically categorizing existing solutions such as differential privacy, federated learning, and prompt encryption. Furthermore, we analyze vulnerabilities unique to mobile LLMs, including adversarial attacks, membership inference, and side-channel attacks, offering an in-depth comparison of their effectiveness and limitations. Despite recent advancements, mobile LLMs face unique hurdles in achieving robust security while maintaining efficiency in resource-constrained environments. To bridge this gap, we propose potential applications, discuss open challenges, and suggest future research directions, paving the way for the development of trustworthy, privacy-compliant, and scalable mobile LLM systems.

Figures

Figures reproduced from arXiv: 2509.02411 by the authors.

Figure 1
Figure 1. Overview of This Survey: Towards Privacy and Security in Mobile Large Language Models [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Data Flow in LLM’s Architecture LLM Architecture 1) Model Compression: Model compression techniques are designed to reduce the size and complexity of LLMs while preserving as much of their original performance as possible. Key strategies include: (1) Pruning removes less critical parts of the model, effectively reducing the number of parameters that need processing, thereby streamlining the model [30], [31]. (2) Qua… view at source ↗
Figure 3
Figure 3. Collaborative Edge Computing with Cloud LLMs [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Membership Inference Attack on Mobile LLMs [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 8
Figure 8. Figure 8: Trustworthy Mobile LLMs in Different Applications [PITH_FULL_IMAGE:figures/full_fig_p006_8.png]
Figure 7
Figure 7. Figure 7: Side-Channel Attack on Mobile LLMs VI. TRUSTWORTHY MOBILE LLMS PRACTICES In this section, we conduct a thorough examination of main￾stream industries, including healthcare, finance, and education, that have successfully implemented LLMs while maintaining data security …

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

92 extracted references · 49 canonical work pages

  1. [80]

    Llm-fin: Large language models fingerprinting attack on edge devices,

    N. Nazari, F. Xiang, C. Fang, H. M. Makrani, A. Puri, K. Patwari, H. Sayadi, S. Rafatirad, C.-N. Chuah, and H. Homayoun, “Llm-fin: Large language models fingerprinting attack on edge devices,” in 2024 25th International Symposium on Quality Electronic Design (ISQED) . IEEE, 2024, pp. 1–6

    work page 2024

  2. [1]

    From llms to llm- based agents for software engineering: A survey of current, challenges and future,

    H. Jin, L. Huang, H. Cai, J. Yan, B. Li, and H. Chen, “From llms to llm- based agents for software engineering: A survey of current, challenges and future,” arXiv preprint arXiv:2408.02479 , 2024

    arXiv 2024

  3. [2]

    On-device language models: A comprehensive review,

    J. Xu, Z. Li, W. Chen, Q. Wang, X. Gao, Q. Cai, and Z. Ling, “On-device language models: A comprehensive review,” arXiv preprint arXiv:2409.00088, 2024

    arXiv 2024

  4. [3]

    Provably efficient service function chain em- bedding and protection in edge networks,

    D. Zheng and X. Cao, “Provably efficient service function chain em- bedding and protection in edge networks,” IEEE/ACM Trans. on Net. , pp. 1–16, 2024

    2024

  5. [4]

    Enhancing hipaa compliance in ai-driven mhealth devices security and privacy,

    A. K. I. Riad, M. A. Barek, M. M. Rahman, M. S. Akter, T. Islam, M. A. Rahman, M. R. Mia, H. Shahriar, F. Wu, and S. I. Ahamed, “Enhancing hipaa compliance in ai-driven mhealth devices security and privacy,” in 2024 IEEE 48th Annual Computers, Software, and Applications Conference (COMPSAC). IEEE, 2024, pp. 2430–2435

    2024

  6. [5]

    Revolutionizing finance with llms: An overview of applications and insights,

    H. Zhao, Z. Liu, Z. Wu, Y . Li, T. Yang, P. Shu, S. Xu, H. Dai, L. Zhao, G. Mai et al. , “Revolutionizing finance with llms: An overview of applications and insights,” arXiv preprint arXiv:2401.11641 , 2024

    arXiv 2024

  7. [6]

    Analyzing the impact of a structured llm workshop in different education levels

    V . Kozov, B. Ivanova, K. Shoylekova, and M. Andreeva, “Analyzing the impact of a structured llm workshop in different education levels.” Applied Sciences (2076-3417) , vol. 14, no. 14, 2024

  8. [7]

    Sok: Model reverse engineering threats for neural network hardware,

    S. Potluri and F. Koushanfar, “Sok: Model reverse engineering threats for neural network hardware,” Cryptology ePrint Archive , 2024

    2024

Show all 92 references
  1. [8]

    On protecting the data privacy of large language models (llms): A survey,

    B. Yan, K. Li, M. Xu, Y . Dong, Y . Zhang, Z. Ren, and X. Cheng, “On protecting the data privacy of large language models (llms): A survey,” arXiv preprint arXiv:2403.05156 , 2024

    2024 arXiv

  2. [9]

    Adversarial attacks on large language models,

    J. Zou, S. Zhang, and M. Qiu, “Adversarial attacks on large language models,” in International Conference on Knowledge Science, Engineer- ing and Management . Springer, 2024, pp. 85–96

    2024

  3. [10]

    Mapping your model: Assessing the impact of adversarial attacks on llm-based programming assistants,

    J. Heibel and D. Lowd, “Mapping your model: Assessing the impact of adversarial attacks on llm-based programming assistants,” arXiv preprint arXiv:2407.11072, 2024

    2024 arXiv

  4. [11]

    Not what you’ve signed up for: Compromising real-world llm- integrated applications with indirect prompt injection,

    K. Greshake, S. Abdelnabi, S. Mishra, C. Endres, T. Holz, and M. Fritz, “Not what you’ve signed up for: Compromising real-world llm- integrated applications with indirect prompt injection,” in Proceedings of the 16th ACM Workshop on Artificial Intelligence and Security , 2023,...

    2023

  5. [12]

    Do membership inference attacks work on large language models?

    M. Duan, A. Suri, N. Mireshghallah, S. Min, W. Shi, L. Zettlemoyer, Y . Tsvetkov, Y . Choi, D. Evans, and H. Hajishirzi, “Do membership inference attacks work on large language models?” arXiv preprint arXiv:2402.07841, 2024

    2024 arXiv

  6. [13]

    Llms and the future of chip design: Unveiling security risks and building trust,

    Z. Wang, L. Alrahis, L. Mankali, J. Knechtel, and O. Sinanoglu, “Llms and the future of chip design: Unveiling security risks and building trust,” arXiv preprint arXiv:2405.07061 , 2024

    2024 arXiv

  7. [14]

    Anonymizing medical documents with local, privacy preserving large language models: The llm-anonymizer,

    I. C. Wiest, M.-E. Lessmann, F. Wolf, D. Ferber, M. Van Treeck, J. Zhu, M. P. Ebert, C. B. Westphalen, M. Wermke, and J. N. Kather, “Anonymizing medical documents with local, privacy preserving large language models: The llm-anonymizer,” medRxiv, pp. 2024–06, 2024

    2024

  8. [15]

    Incognitext: Privacy-enhancing conditional text anonymization via llm- based private attribute randomization,

    A. Frikha, N. Walha, K. K. Nakka, R. Mendes, X. Jiang, and X. Zhou, “Incognitext: Privacy-enhancing conditional text anonymization via llm- based private attribute randomization,”arXiv preprint arXiv:2407.02956, 2024

    2024 arXiv

  9. [16]

    Whispered tuning: Data privacy preservation in fine-tuning llms through differential privacy,

    T. Singh, H. Aditya, V . K. Madisetti, and A. Bahga, “Whispered tuning: Data privacy preservation in fine-tuning llms through differential privacy,” Journal of Software Engineering and Applications , vol. 17, no. 1, pp. 1–22, 2024

    2024

  10. [17]

    Promptcrypt: Prompt encryption for secure communication with large language models,

    G. Lin, W. Hua, and Y . Zhang, “Promptcrypt: Prompt encryption for secure communication with large language models,” arXiv preprint arXiv:2402.05868, 2024

    2024 arXiv

  11. [18]

    Llms can understand encrypted prompt: Towards privacy-computing friendly transformers,

    X. Liu and Z. Liu, “Llms can understand encrypted prompt: Towards privacy-computing friendly transformers,” arXiv preprint arXiv:2305.18396, 2023

    2023 arXiv

  12. [19]

    Large language models: a comprehensive survey of its applications, challenges, limitations, and future prospects,

    M. U. Hadi, Q. Al Tashi, A. Shah, R. Qureshi, A. Muneer, M. Irfan, A. Zafar, M. B. Shaikh, N. Akhtar, J. Wuet al., “Large language models: a comprehensive survey of its applications, challenges, limitations, and future prospects,” Authorea Preprints, 2024

    2024

  13. [20]

    Recent advances in gen- erative ai and large language models: Current status, challenges, and perspectives,

    D. H. Hagos, R. Battle, and D. B. Rawat, “Recent advances in gen- erative ai and large language models: Current status, challenges, and perspectives,” IEEE Transactions on Artificial Intelligence , 2024

    2024

  14. [21]

    A review on large language models: Architectures, applications, taxonomies, open issues and challenges,

    M. A. K. Raiaan, M. S. H. Mukta, K. Fatema, N. M. Fahad, S. Sakib, M. M. J. Mim, J. Ahmad, M. E. Ali, and S. Azam, “A review on large language models: Architectures, applications, taxonomies, open issues and challenges,” IEEE Access, 2024

    2024

  15. [22]

    Harnessing the power of llms in practice: A survey on chatgpt and beyond,

    J. Yang, H. Jin, R. Tang, X. Han, Q. Feng, H. Jiang, S. Zhong, B. Yin, and X. Hu, “Harnessing the power of llms in practice: A survey on chatgpt and beyond,” ACM Transactions on Knowledge Discovery from Data, vol. 18, no. 6, pp. 1–32, 2024

    2024

  16. [23]

    Llms: Evolution and new frontiers,

    U. Kamath, K. Keenan, G. Somers, and S. Sorenson, “Llms: Evolution and new frontiers,” in Large Language Models: A Deep Dive: Bridging Theory and Practice . Springer, 2024, pp. 423–438

    2024

  17. [24]

    A novel time–frequency transformer based on self–attention mechanism and its application in fault diagnosis of rolling bearings,

    Y . Ding, M. Jia, Q. Miao, and Y . Cao, “A novel time–frequency transformer based on self–attention mechanism and its application in fault diagnosis of rolling bearings,” Mechanical Systems and Signal Processing, vol. 168, p. 108616, 2022

    2022

  18. [25]

    Context-aware self- attention networks for natural language processing,

    B. Yang, L. Wang, D. F. Wong, S. Shi, and Z. Tu, “Context-aware self- attention networks for natural language processing,” Neurocomputing, vol. 458, pp. 157–169, 2021

    2021

  19. [26]

    A review of recurrent neural networks: Lstm cells and network architectures,

    Y . Yu, X. Si, C. Hu, and J. Zhang, “A review of recurrent neural networks: Lstm cells and network architectures,” Neural computation , vol. 31, no. 7, pp. 1235–1270, 2019

    2019

  20. [27]

    Understanding llms: A comprehensive overview from training to inference,

    Y . Liu, H. He, T. Han, X. Zhang, M. Liu, J. Tian, Y . Zhang, J. Wang, X. Gao, T. Zhong et al. , “Understanding llms: A comprehensive overview from training to inference,” arXiv preprint arXiv:2401.02038 , 2024

    2024 arXiv

  21. [28]

    The training process and methods for llms using an own knowledge base,

    S. Zhiyuan et al., “The training process and methods for llms using an own knowledge base,” Journal of Artificial Intelligence Practice , vol. 7, no. 3, pp. 41–47, 2024

    2024

  22. [29]

    Fine-tuning and utilization methods of domain-specific llms,

    C. Jeong, “Fine-tuning and utilization methods of domain-specific llms,” arXiv preprint arXiv:2401.02981 , 2024

    2024 arXiv

  23. [30]

    Llm-pruner: On the structural pruning of large language models,

    X. Ma, G. Fang, and X. Wang, “Llm-pruner: On the structural pruning of large language models,” Advances in neural information processing systems, vol. 36, pp. 21 702–21 720, 2023

    2023

  24. [31]

    Mini-llm: Memory-efficient structured pruning for large language models,

    H. Cheng, M. Zhang, and J. Q. Shi, “Mini-llm: Memory-efficient structured pruning for large language models,” arXiv preprint arXiv:2407.11681, 2024

    2024 arXiv

  25. [32]

    I- llm: Efficient integer-only inference for fully-quantized low-bit large language models,

    X. Hu, Y . Chen, D. Yang, S. Zhou, Z. Yuan, J. Yu, and C. Xu, “I- llm: Efficient integer-only inference for fully-quantized low-bit large language models,” arXiv preprint arXiv:2405.17849 , 2024

    2024 arXiv

  26. [33]

    Slim-llm: Salience-driven mixed-precision quantization for large language models,

    W. Huang, H. Qin, Y . Liu, Y . Li, X. Liu, L. Benini, M. Magno, and X. Qi, “Slim-llm: Salience-driven mixed-precision quantization for large language models,” arXiv preprint arXiv:2405.14917 , 2024

    2024 arXiv

  27. [34]

    Efficient llm context distillation,

    R. Upadhayayaya, Z. Smith, C. Kottmyer, and M. R. Osti, “Efficient llm context distillation,” arXiv preprint arXiv:2409.01930 , 2024

    2024

  28. [35]

    Reducing llm hallucination using knowledge distillation: A case study with mistral large and mmlu benchmark,

    D. McDonald, R. Papadopoulos, and L. Benningfield, “Reducing llm hallucination using knowledge distillation: A case study with mistral large and mmlu benchmark,” Authorea Preprints, 2024

    2024

  29. [36]

    Edgeshard: Efficient llm inference via collaborative edge computing,

    M. Zhang, J. Cao, X. Shen, and Z. Cui, “Edgeshard: Efficient llm inference via collaborative edge computing,” arXiv preprint arXiv:2405.14371, 2024

    2024 arXiv

  30. [37]

    Empowering generative ai through mobile edge computing,

    L. Ale, N. Zhang, S. A. King, and D. Chen, “Empowering generative ai through mobile edge computing,” Nature Reviews Electrical Engineer- ing, pp. 1–9, 2024

    2024

  31. [38]

    Iotdq: An industrial iot data analysis library for apache iotdb,

    P. Chen, W. He, W. Ma, X. Huang, and C. Wang, “Iotdq: An industrial iot data analysis library for apache iotdb,” Big Data Min. Anal. , vol. 7, no. 1, pp. 29–41, 2024

    2024

  32. [39]

    Data protection in digital learning space: An overview,

    E. Djeki, J. D ´egila, C. Bondiombouy, and M. H. Alhassan, “Data protection in digital learning space: An overview,” in AIP Conference Proceedings, vol. 3109, no. 1. AIP Publishing, 2024

    2024

  33. [40]

    In- formation disclosure on mobile devices: Re-examining privacy calculus with actual user behavior,

    M. J. Keith, S. C. Thompson, J. Hale, P. B. Lowry, and C. Greer, “In- formation disclosure on mobile devices: Re-examining privacy calculus with actual user behavior,” International journal of human-computer studies, vol. 71, no. 12, pp. 1163–1173, 2013

    2013

  34. [41]

    Data security and privacy-preserving in edge computing paradigm: Survey and open issues,

    J. Zhang, B. Chen, Y . Zhao, X. Cheng, and F. Hu, “Data security and privacy-preserving in edge computing paradigm: Survey and open issues,” IEEE access, vol. 6, pp. 18 209–18 237, 2018

    2018

  35. [42]

    When llms meet cybersecurity: A systematic literature review,

    J. Zhang, H. Bu, H. Wen, Y . Chen, L. Li, and H. Zhu, “When llms meet cybersecurity: A systematic literature review,” arXiv preprint arXiv:2405.03644, 2024

    2024 arXiv

  36. [43]

    The emerged security and privacy of llm agent: A survey with case studies,

    F. He, T. Zhu, D. Ye, B. Liu, W. Zhou, and P. S. Yu, “The emerged security and privacy of llm agent: A survey with case studies,” arXiv preprint arXiv:2407.19354, 2024. IEEE INTERNET OF THINGS JOURNAL, V ol. XX, No. XX 9

    2024

  37. [44]

    Llm-pbe: Assessing data privacy in large language models,

    Q. Li, J. Hong, C. Xie, J. Tan, R. Xin, J. Hou, X. Yin, Z. Wang, D. Hendrycks, Z. Wang et al. , “Llm-pbe: Assessing data privacy in large language models,” arXiv preprint arXiv:2408.12787 , 2024

    2024 arXiv

  38. [45]

    Task-specific knowledge distillation with differen- tial privacy in llms,

    S. Garg and V . Torra, “Task-specific knowledge distillation with differen- tial privacy in llms,” in European Symposium on Research in Computer Security. Springer, 2024, pp. 374–389

    2024

  39. [46]

    Multi-smart meter data encryption scheme based on distributed differential privacy,

    R. Yan, Y . Zheng, N. Yu, and C. Liang, “Multi-smart meter data encryption scheme based on distributed differential privacy,” Big Data Min. Anal., vol. 7, no. 1, pp. 131–141, 2024

    2024

  40. [47]

    Openfedllm: Training large language models on decentralized private data via federated learning,

    R. Ye, W. Wang, J. Chai, D. Li, Z. Li, Y . Xu, Y . Du, Y . Wang, and S. Chen, “Openfedllm: Training large language models on decentralized private data via federated learning,” in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , 2024, pp. ...

    2024

  41. [48]

    Federatedscope-llm: A comprehensive package for fine-tuning large language models in federated learning,

    W. Kuang, B. Qian, Z. Li, D. Chen, D. Gao, X. Pan, Y . Xie, Y . Li, B. Ding, and J. Zhou, “Federatedscope-llm: A comprehensive package for fine-tuning large language models in federated learning,” in Proceed- ings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Da...

    2024

  42. [49]

    Fedstr: Money-in ai-out— a decentralized marketplace for federated learning and llm training on the nostr protocol,

    K. E. Nikolakakis, G. Chantzialexiou, and D. Kalogerias, “Fedstr: Money-in ai-out— a decentralized marketplace for federated learning and llm training on the nostr protocol,”arXiv preprint arXiv:2404.15834, 2024

    2024

  43. [50]

    Fedbert: When federated learning meets pre-training,

    Y . Tian, Y . Wan, L. Lyu, D. Yao, H. Jin, and L. Sun, “Fedbert: When federated learning meets pre-training,” ACM Trans. Intell. Syst. Technol., vol. 13, no. 4, 2022

    2022

  44. [51]

    FedPETuning: When federated learning meets the parameter-efficient tuning methods of pre-trained language models,

    Z. Zhang, Y . Yang, Y . Dai, Q. Wang, Y . Yu, L. Qu, and Z. Xu, “FedPETuning: When federated learning meets the parameter-efficient tuning methods of pre-trained language models,” in Findings of the As- sociation for Computational Linguistics: ACL 2023 . Toronto, Canada: Assoc...

    2023

  45. [52]

    Towards building the federatedgpt: Federated instruction tun- ing,

    J. Zhang, S. Vahidian, M. Kuo, C. Li, R. Zhang, T. Yu, G. Wang, and Y . Chen, “Towards building the federatedgpt: Federated instruction tun- ing,” in ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2024, pp. 6915–6919

    2024

  46. [53]

    Fedpit: Towards privacy-preserving and few-shot federated instruction tuning,

    Z. Zhang, J. Zhang, J. Huang, L. Qu, H. Zhang, and Z. Xu, “Fedpit: Towards privacy-preserving and few-shot federated instruction tuning,” arXiv preprint arXiv:2403.06131 , 2024

    2024 arXiv

  47. [54]

    Flora: Federated fine-tuning large language models with heterogeneous low- rank adaptations,

    Z. Wang, Z. Shen, Y . He, G. Sun, H. Wang, L. Lyu, and A. Li, “Flora: Federated fine-tuning large language models with heterogeneous low- rank adaptations,” arXiv preprint arXiv:2409.05976 , 2024

    2024 arXiv

  48. [55]

    Fedbiot: Llm local fine-tuning in federated learning without full model,

    F. Wu, Z. Li, Y . Li, B. Ding, and J. Gao, “Fedbiot: Llm local fine-tuning in federated learning without full model,” in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , 2024, pp. 3345–3355

    2024

  49. [56]

    LoRA: Low-rank adaptation of large language models,

    E. J. Hu, yelong shen, P. Wallis, Z. Allen-Zhu, Y . Li, S. Wang, L. Wang, and W. Chen, “LoRA: Low-rank adaptation of large language models,” in International Conference on Learning Representations , 2022. [Online]. Available: https://openreview.net/forum?id=nZeVKeeFYf9

    2022

  50. [57]

    Federated lora with sparse communication,

    K. Kuo, A. Raje, K. Rajesh, and V . Smith, “Federated lora with sparse communication,” arXiv preprint arXiv:2406.05233 , 2024

    2024 arXiv

  51. [58]

    SplitLoRA: A Split Parameter-Efficient Fine-Tuning Framework for Large Language Models,

    Z. Lin, X. Hu, Y . Zhang, Z. Chen, Z. Fang, X. Chen, A. Li, P. Vepakomma, and Y . Gao, “SplitLoRA: A Split Parameter-Efficient Fine-Tuning Framework for Large Language Models,” arXiv preprint arXiv:2407.00952, 2024

    2024 arXiv

  52. [59]

    Federa: Efficient fine-tuning of language models in federated learning leveraging weight decomposi- tion,

    Y . Yan, Q. Yang, S. Tang, and Z. Shi, “Federa: Efficient fine-tuning of language models in federated learning leveraging weight decomposi- tion,” arXiv preprint arXiv:2404.18848 , 2024

    2024 arXiv

  53. [60]

    Federated fine-tuning of large language models under heterogeneous tasks and client resources,

    J. Bai, D. Chen, B. Qian, L. Yao, and Y . Li, “Federated fine-tuning of large language models under heterogeneous tasks and client resources,” in The Thirty-eighth Annual Conference on Neural Information Process- ing Systems, 2024

    2024

  54. [61]

    Adversarial evasion on llms,

    R. Guerraoui and R. Pinot, “Adversarial evasion on llms,” Large, p. 181, 2024

    2024

  55. [62]

    Adversarial evasion attack efficiency against large language models,

    J. Vitorino, E. Maia, and I. Prac ¸a, “Adversarial evasion attack efficiency against large language models,” arXiv preprint arXiv:2406.08050, 2024

    2024 arXiv

  56. [63]

    Human- imperceptible retrieval poisoning attacks in llm-powered applications,

    Q. Zhang, B. Zeng, C. Zhou, G. Go, H. Shi, and Y . Jiang, “Human- imperceptible retrieval poisoning attacks in llm-powered applications,” in Companion Proceedings of the 32nd ACM International Conference on the Foundations of Software Engineering , 2024, pp. 502–506

    2024

  57. [64]

    Instruction backdoor attacks against customized {LLMs},

    R. Zhang, H. Li, R. Wen, W. Jiang, Y . Zhang, M. Backes, Y . Shen, and Y . Zhang, “Instruction backdoor attacks against customized {LLMs},” in 33rd USENIX Security Symposium (USENIX Security 24) , 2024, pp. 1849–1866

    2024

  58. [65]

    Towards resilient and efficient llms: A comparative study of efficiency, performance, and adversarial robustness,

    X. Fan and C. Tao, “Towards resilient and efficient llms: A comparative study of efficiency, performance, and adversarial robustness,” arXiv preprint arXiv:2408.04585, 2024

    2024 arXiv

  59. [66]

    Enhancing adversarial robustness of llms with analytic hierarchy process,

    J. Zhao, M. Wang, N. Xu, W. Mao et al. , “Enhancing adversarial robustness of llms with analytic hierarchy process,” in First Conference on Language Modeling , 2024

    2024

  60. [67]

    Forget and rewire: Enhancing the resilience of transformer-based models against {Bit-Flip} attacks,

    N. Nazari, H. M. Makrani, C. Fang, H. Sayadi, S. Rafatirad, K. N. Khasawneh, and H. Homayoun, “Forget and rewire: Enhancing the resilience of transformer-based models against {Bit-Flip} attacks,” in 33rd USENIX Security Symposium (USENIX Security 24) , 2024, pp. 1349–1366

    2024

  61. [68]

    Understanding generalization and robustness in modern deep learning,

    M. Andriushchenko, “Understanding generalization and robustness in modern deep learning,” EPFL, Tech. Rep., 2024

    2024

  62. [69]

    Noisy neighbors: Efficient member- ship inference attacks against llms,

    F. Galli, L. Melis, and T. Cucinotta, “Noisy neighbors: Efficient member- ship inference attacks against llms,” arXiv preprint arXiv:2406.16565 , 2024

    2024 arXiv

  63. [70]

    Why does differential privacy with large epsilon defend against practical membership inference attacks?

    A. Lowy, Z. Li, J. Liu, T. Koike-Akino, K. Parsons, and Y . Wang, “Why does differential privacy with large epsilon defend against practical membership inference attacks?” arXiv preprint arXiv:2402.09540, 2024

    2024 arXiv

  64. [71]

    Ew- tune: A framework for privately fine-tuning large language models with differential privacy,

    R. Behnia, M. R. Ebrahimi, J. Pacheco, and B. Padmanabhan, “Ew- tune: A framework for privately fine-tuning large language models with differential privacy,” in 2022 IEEE International Conference on Data Mining Workshops (ICDMW). IEEE, 2022, pp. 560–566

    2022

  65. [72]

    Practical membership inference attacks against fine-tuned large language models via self-prompt calibration,

    W. Fu, H. Wang, C. Gao, G. Liu, Y . Li, and T. Jiang, “Practical membership inference attacks against fine-tuned large language models via self-prompt calibration,” arXiv preprint arXiv:2311.06062 , 2023

    2023 arXiv

  66. [73]

    Protecting llms against privacy attacks while preserving utility,

    G. Dhingra, S. Sood, Z. M. Wase, A. Bahga, and V . K. Madisetti, “Protecting llms against privacy attacks while preserving utility,”Journal of Information Security , vol. 15, no. 4, pp. 448–473, 2024

    2024

  67. [74]

    Covert malicious finetuning: Challenges in safeguarding llm adaptation,

    D. Halawi, A. Wei, E. Wallace, T. T. Wang, N. Haghtalab, and J. Steinhardt, “Covert malicious finetuning: Challenges in safeguarding llm adaptation,” arXiv preprint arXiv:2406.20053 , 2024

    2024 arXiv

  68. [75]

    Text embedding inversion security for multilingual language models,

    Y . Chen, H. Lent, and J. Bjerva, “Text embedding inversion security for multilingual language models,” in Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2024, pp. 7808–7827

    2024

  69. [76]

    Transferable embedding inversion attack: Uncovering privacy risks in text embed- dings without model queries,

    Y .-H. Huang, Y . Tsai, H. Hsiao, H.-Y . Lin, and S.-D. Lin, “Transferable embedding inversion attack: Uncovering privacy risks in text embed- dings without model queries,” arXiv preprint arXiv:2406.10280 , 2024

    2024 arXiv

  70. [77]

    Robust and private federated learning on llms,

    R. Guerraoui and N. Gupta, “Robust and private federated learning on llms,” Large, p. 189, 2024

    2024

  71. [78]

    Local and central differential privacy for robustness and privacy in federated learning,

    M. Naseri, J. Hayes, and E. De Cristofaro, “Local and central differential privacy for robustness and privacy in federated learning,” arXiv preprint arXiv:2009.03561, 2020

    2009 arXiv

  72. [79]

    Analog-to-digital converters for secure and emerging aiot applications,

    R. Chen, “Analog-to-digital converters for secure and emerging aiot applications,” Ph.D. dissertation, Massachusetts Institute of Technology, 2023

    2023

  73. [81]

    Llmaas: Serving large language models on trusted serverless computing platforms,

    Z. Cai, R. Ma, Y . Fu, W. Zhang, R. Ma, and H. Guan, “Llmaas: Serving large language models on trusted serverless computing platforms,” IEEE Transactions on Artificial Intelligence , 2024

    2024

  74. [82]

    Assessing healthcare software built using iot and llm technologies,

    G. De Vito, “Assessing healthcare software built using iot and llm technologies,” in Proceedings of the 28th International Conference on Evaluation and Assessment in Software Engineering, 2024, pp. 476–481

    2024

  75. [83]

    Integrating large language models for enhanced clinical decision support systems in modern healthcare,

    K. J. Prabhod, “Integrating large language models for enhanced clinical decision support systems in modern healthcare,” Journal of Machine Learning for Healthcare Decision Support , vol. 3, no. 1, pp. 18–62, 2023

    2023

  76. [84]

    Differential privacy preserving in big data analytics for connected health,

    C. Lin, Z. Song, H. Song, Y . Zhou, Y . Wang, and G. Wu, “Differential privacy preserving in big data analytics for connected health,” Journal of medical systems , vol. 40, pp. 1–9, 2016

    2016

  77. [85]

    Add noise to remove noise: Local differential privacy for feature selection,

    M. Alishahi, V . Moghtadaiee, and H. Navidan, “Add noise to remove noise: Local differential privacy for feature selection,” Computers & Security, vol. 123, p. 102934, 2022

    2022

  78. [86]

    Assessing the potential integration of large language models in accounting practices: evidence from an emerging economy,

    A. A. Toumeh, “Assessing the potential integration of large language models in accounting practices: evidence from an emerging economy,” Future Business Journal , vol. 10, no. 1, p. 82, 2024

    2024

  79. [87]

    Intelligent agents with llm-based process automation,

    Y . Guan, D. Wang, Z. Chu, S. Wang, F. Ni, R. Song, and C. Zhuang, “Intelligent agents with llm-based process automation,” in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024, pp. 5018–5027

    2024

  80. [88]

    Evaluating privacy compli- ance in commercial large language models-chatgpt, claude, and gemini,

    O. Cartwright, H. Dunbar, and T. Radcliffe, “Evaluating privacy compli- ance in commercial large language models-chatgpt, claude, and gemini,” 2024. IEEE INTERNET OF THINGS JOURNAL, V ol. XX, No. XX 10

    2024

  81. [89]

    A trustworthy service transaction framework for privacy protection,

    Z. Li, T. Mo, W. Li, and Z. Tu, “A trustworthy service transaction framework for privacy protection,” in International Conference on Service Science. Springer, 2024, pp. 107–121

    2024

  82. [90]

    Towards effective teaching assistants: From intent-based chatbots to llm-powered teaching assistants,

    B. Alsafari, E. Atwell, A. Walker, and M. Callaghan, “Towards effective teaching assistants: From intent-based chatbots to llm-powered teaching assistants,” Natural Language Processing Journal , p. 100101, 2024

    2024

  83. [91]

    Enabling on-device llms personalization with smartphone sensing,

    S. Zhang, Y . Ma, L. Fang, H. Jia, S. D’Alfonso, and V . Kostakos, “Enabling on-device llms personalization with smartphone sensing,” arXiv preprint arXiv:2407.04418 , 2024

    2024 arXiv

  84. [92]

    Privacy-preserving learning analytics: challenges and techniques,

    M. E. Gursoy, A. Inan, M. E. Nergiz, and Y . Saygin, “Privacy-preserving learning analytics: challenges and techniques,” IEEE Transactions on Learning technologies, vol. 10, no. 1, pp. 68–81, 2016

    2016

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.