Pith. sign in

REVIEW 5 major objections 5 minor 1 cited by

SafeSynthDP: Leveraging Large Language Models for Privacy-Preserving Synthetic Data Generation Using Differential Privacy

T0 review · 5 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The paper proposes a training-free pipeline that generates synthetic text with an LLM, perturbs its word frequencies with Laplace or Gaussian noise, and claims this produces differentially private data that still supports classification…

desk verdict A clearly written paper whose central privacy claim is unsupported: noise added after LLM generation cannot provide differential privacy, and no mechanism or MIA evaluation is ever defined. read the letter →

arxiv 2412.20641 v1 pith:45MFGYXM submitted 2024-12-30 cs.LG cs.CR

Md Mahadi Hasan Nahid , Sadid Bin Hasan This is my paper · ORCID
classification cs.LGcs.CR
keywords differentialprivacysyntheticdatalargelanguagemodelsin-contextlearningtextclassificationLaplacemechanismGaussianprivacy-utilitytrade-off
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper proposes SafeSynthDP, a training-free pipeline that uses a large language model to generate a synthetic version of a sensitive text dataset and then perturbs word frequencies with Laplace or Gaussian noise, claiming this yields differentially private synthetic data. The authors evaluate the approach on the AGNews news-classification task, training four machine learning models and testing two LLMs in zero-, 2-, and 4-shot in-context learning settings. Their experiments show that models trained or prompted with the noisy synthetic data retain most of their accuracy, with simple classifiers losing roughly 3 to 10 percentage points and LLM in-context learning narrowing the gap when given more demonstrations. The paper's central claim is that this integration offers a viable balance between privacy protection and data utility, providing a foundation for compliant synthetic data use in sensitive domains.

What carries the argument

The central mechanism is the SafeSynthDP pipeline: a small set of labeled examples from the original dataset is placed in the prompt, the LLM generates a synthetic corpus in JSON form, and then Laplace or Gaussian noise is added to the token frequencies of the generated text, with the privacy budget epsilon controlling how much noise is applied. The in-context demonstrations carry the statistical style of the original data, while the post-hoc noise is what the paper relies on for privacy protection. The evaluation then measures how well each model type tolerates that noise.

What would settle it

Measure the change in the synthetic output distribution when a single example is removed from the in-context prompt; if the change is larger than the claimed epsilon would permit under any plausible sensitivity bound, the differential privacy claim is false.

Watch

Extended reading notes

Core claim

In the paper's own terms, the discovery is that differentially private noise can be injected into the word-frequency distribution of LLM-generated text without destroying its usefulness for text classification. Using the AGNews dataset and gpt-4o-mini as the generator, the authors report that Multinomial Naive Bayes and SVM trained on noisy synthetic data reach 77.92% and 76.43% accuracy respectively, compared with 80.73% and 86.75% on the original data; the recurrent models drop further, to about 65%. For in-context learning, both gpt-4o-mini and gemini-1.5-flash perform better with 4-shot demonstrations than with 2-shot, and the gap between original and synthetic data shrinks as shot count increases, supporting the claim that privacy-enhanced synthetic data can substitute for real data in some settings.

Load-bearing premise

The load-bearing premise is that adding Laplace or Gaussian noise to the word frequencies of an already-generated synthetic text provides a differential privacy guarantee for the original dataset, but the paper never defines a sensitivity bound, a specific DP mechanism, or a privacy accounting for that step.

Editorial extensions

If this is right

  • Simple frequency-based classifiers (MNB, SVM) lose little accuracy on privacy-enhanced synthetic data, suggesting the method suits high-level statistical tasks.
  • Sequence models (GRU, LSTM) are more sensitive to the noise, indicating that tasks requiring temporal or semantic nuance will need better noise calibration.
  • Increasing the number of in-context demonstrations narrows the performance gap between original and synthetic data for LLM classification.
  • Tuning the privacy budget epsilon provides a practical lever for trading utility against privacy, with epsilon=10 giving the best accuracy in their ICL experiments.
  • The approach is training-free, so it can generate private synthetic data without retraining a generative model for each new dataset.

Reading between the lines

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

  • The paper does not define a formal DP mechanism: no sensitivity bound, no privacy accounting, and the noise is applied after generation, so the guarantee for the original data is not established by the experiments shown.
  • A testable extension would be to measure the actual sensitivity of the in-context generation step by removing one prompt example and estimating the change in the output distribution.
  • The method could be strengthened by applying DP during generation (e.g., to the prompt selection or the sampling step) rather than only to the generated text's token frequencies.
  • Membership inference evaluation, which the paper explicitly lists as future work, would be the natural direct check of the privacy claim.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 5 minor

Summary. The paper proposes SafeSynthDP, a training-free pipeline that uses gpt-4o-mini with in-context examples from a private dataset to generate synthetic text, then injects Laplace or Gaussian noise into token frequencies of the generated text, claiming this yields differentially private synthetic data. The utility of the resulting synthetic data is evaluated by training MNB, SVM, GRU, and LSTM classifiers and by using the synthetic data as demonstrations in in-context learning with gpt-4o-mini and gemini-1.5-flash, all on a sampled subset of AGNews. The reported results show moderate accuracy losses relative to original data, and the paper interprets these as evidence of a viable privacy-utility trade-off controlled by a privacy budget epsilon.

Significance. If the central privacy claim were established, the paper would offer a lightweight, LLM-based alternative to model-centric DP synthetic data generation, with potential value for text classification in privacy-sensitive domains. The paper does provide a concrete evaluation across multiple model families and two LLMs, and it describes its prompts and hyperparameters in appendices, which aids reproducibility of the utility experiments. However, the entire contribution hinges on the assertion that the described noise injection provides differential privacy for the original dataset, and that assertion is not supported by any mechanism definition, sensitivity analysis, or privacy accounting. The absence of the promised membership-inference evaluation and the lack of any statistical uncertainty in the results further weaken the empirical claims. The paper therefore cannot currently be considered a valid demonstration of privacy-preserving synthetic data generation.

major comments (5)
  1. [Section 3.1, "Privacy Enhancement through DP Mechanisms"] The described pipeline feeds in-context examples from the original dataset to gpt-4o-mini, generates text, and only then adds Laplace or Gaussian noise to token frequencies in the generated text. Differential privacy is a property of a randomized algorithm whose input is the original dataset; post-processing can only preserve privacy if the preceding step is already private. Here, the LLM generation step directly used sensitive examples, so no epsilon-DP guarantee for the original dataset follows from adding noise to already-generated text. The paper never defines the query, its L1/L2 sensitivity, or the noise calibration that would connect the reported epsilon values to a formal DP guarantee, and Section 5.4 does not supply this missing analysis instead of asserting it.
  2. [Table 3 and Section 5.3] The row labeled "LLM ICL (epsilon = 0)" reports an accuracy of 69.83%. Under the standard Laplace or Gaussian DP mechanism, epsilon = 0 corresponds to perfect privacy and hence infinite noise, which would drive accuracy toward random chance (approximately 25% for four classes). A near-normal accuracy at epsilon = 0 either indicates that the privacy budget was not implemented as described or that the epsilon values are labels not attached to any actual DP mechanism. This is a load-bearing inconsistency because the claimed privacy-utility trade-off is the paper's central empirical contribution.
  3. [Abstract and Section 5.4] The abstract states that the paper assesses "the resilience of the generated synthetic data to membership inference attacks and related threats," and Section 5.4 asserts that the noise protects against membership inference, but no membership-inference experiment appears anywhere in the manuscript. The absence of this promised evaluation is critical because the privacy claim rests entirely on the assertion that noise prevents membership inference, and that assertion is tested nowhere.
  4. [Tables 1, 2, 3 and Section 5] All reported accuracies are single-run point estimates with no standard deviations, no multiple seeds, and no statistical significance testing. Given the small scope (one dataset, 12,000 training samples), the differences between original and synthetic data and across epsilon values cannot be distinguished from random variation. The quantitative claims about the privacy-utility trade-off are therefore not supported by the reported experiments.
  5. [Limitations (Section 6, final paragraph before References)] The Limitations section explicitly states that "If the noise added is not appropriately calibrated for the dataset or if new, more sophisticated attack methods emerge, the privacy protections might be undermined." This concedes exactly what the paper elsewhere assumes: that the noise is appropriately calibrated. Since no calibration or sensitivity analysis is provided, the paper's own limitation statement reinforces the conclusion that the DP guarantee is not established.
minor comments (5)
  1. [Section 3.1] The sentence "in highly sensitive contexts, we would choose a higher ϵ to maintain data fidelity while still offering privacy protection" reverses the standard convention; higher epsilon means weaker privacy and higher fidelity, not the other way around.
  2. [Section 4.1] The paper states that "we focused on reporting results for the AGNews dataset" but does not describe any other dataset or explain why results for other datasets are omitted; this phrasing should be clarified.
  3. [Section B.3] The prompt text contains the typo "Bussiness" (instead of "Business") and the later ICL prompt contains "follwoing" (instead of "following"); these should be corrected.
  4. [Section 4.4] The interpretation of accuracy above 50% as evidence that the synthetic data "captured some meaningful signals" is not supported by a comparison to a random baseline on the AGNews test set; such a baseline should be reported.
  5. [Section B.4] The ICL prompt shown in the appendix uses synthetic examples with clearly invented headlines (e.g., "Breakthrough in Renewable Energy Technology"), but the main text does not describe how the demonstrations were selected from the synthetic corpus; this selection procedure should be stated.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the core privacy claim is unsupported but is not derived from its own inputs or from load-bearing self-citation.

full rationale

The paper contains no derivation in which a predicted quantity is equal to an input by construction. The central privacy claim — that adding Laplace or Gaussian noise to token frequencies in LLM-generated text yields differential privacy — is not established: the paper never defines the query, its sensitivity, or a privacy accounting, and the noise is applied after the LLM has already received in-context examples from the original data (Sections 3.1 and 5.4). The abstract also promises a membership-inference evaluation that is absent from the reported experiments. These are unsupported-assertion and completeness problems, not circularity: the privacy conclusion is not contained in the stated mechanism by definition, no fitted parameter is renamed as a prediction, and no self-citation is used to carry the load-bearing privacy step. The self-citations (Nahid 2024; Nahid and Rafiei 2024a,b) are background references for LLM capabilities with tabular/text data and do not prop up the central claim. The utility evaluations are ordinary empirical comparisons against original-data baselines, so the derivation chain is not circular; it simply lacks a valid privacy argument. Accordingly, the circularity score is 0.

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

The paper introduces no new physical or conceptual entities. Its free parameters are the privacy budget and noise scale, which are chosen ad hoc. The key assumption, that token-count noise on LLM output constitutes differential privacy, is unproven and likely invalid. The reliance on the LLM not leaking the in-context examples is an additional unverified assumption.

free parameters (3)
  • privacy budget epsilon = 0, 0.5, 1, 10 (swept)
    Chosen by hand to balance privacy and utility in Table 3. There is no principled selection procedure, and the relationship between epsilon and the actual noise added is not defined.
  • noise scale / sensitivity = not specified
    The method claims calibrated noise but never defines sensitivity or noise variance. The noise amount is effectively a free choice made to make the method appear to work.
  • number of in-context demonstrations = 4 (num_shots)
    Chosen for the synthetic data generation prompt (Appendix B.3). The choice is not justified and affects the output distribution. In ICL evaluation, 2-shot and 4-shot are also used.
assumptions (3)
  • domain assumption Adding Laplace or Gaussian noise to token frequencies in LLM-generated text provides epsilon-differential privacy for the original dataset.
    This is the load-bearing premise of Sections 3.1 and 5.4, but it is never proven. No sensitivity analysis connects the noise to the original data, and the statement is incorrect as formulated because perturbing post-hoc generated text does not protect the original data unless the generation itself is DP.
  • domain assumption The in-context examples used to prompt the LLM do not themselves leak sensitive information.
    The method feeds a few original examples to the LLM (Section 3.1). Privacy depends on these examples not being memorized or emitted. The paper does not analyze or bound this leakage.
  • standard math Standard definitions and properties of differential privacy hold as background (Dwork et al., 2014).
    The paper invokes DP as the formal framework, but never formally applies it. The background definition in Appendix A.2 is standard, but no DP mechanism is instantiated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SafeSynthDP: Leveraging Large Language Models for Privacy-Preserving Synthetic Data Generation Using Differential Privacy." pith.science (2026). https://pith.science/paper/45MFGYXM

@misc{pith2026241220641,
  author       = {Pith},
  title        = {Pith review of: SafeSynthDP: Leveraging Large Language Models for Privacy-Preserving Synthetic Data Generation Using Differential Privacy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/45MFGYXM}},
  note         = {Machine review of arXiv:2412.20641}
}
read the original abstract

Machine learning (ML) models frequently rely on training data that may include sensitive or personal information, raising substantial privacy concerns. Legislative frameworks such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA) have necessitated the development of strategies that preserve privacy while maintaining the utility of data. In this paper, we investigate the capability of Large Language Models (LLMs) to generate synthetic datasets integrated with Differential Privacy (DP) mechanisms, thereby enabling data-driven research and model training without direct exposure of sensitive information. Our approach incorporates DP-based noise injection methods, including Laplace and Gaussian distributions, into the data generation process. We then evaluate the utility of these DP-enhanced synthetic datasets by comparing the performance of ML models trained on them against models trained on the original data. To substantiate privacy guarantees, we assess the resilience of the generated synthetic data to membership inference attacks and related threats. The experimental results demonstrate that integrating DP within LLM-driven synthetic data generation offers a viable balance between privacy protection and data utility. This study provides a foundational methodology and insight into the privacy-preserving capabilities of LLMs, paving the way for compliant and effective ML research and applications.

Figures

Figures reproduced from arXiv: 2412.20641 by the authors.

Figure 1
Figure 1. Workflow for Generating Privacy-Preserving Synthetic Data ( [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

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. Towards Privacy-aware Mental Health AI Models: Advances, Challenges, and Opportunities

    cs.CL 2025-02 accept novelty 4.0 of 10

    A survey and position paper mapping privacy threats in mental health AI and recommending a pipeline of anonymization, synthetic data, and differential privacy.

Reference graph

Works this paper leans on

41 extracted references · 18 canonical work pages · cited by 1 Pith paper

  1. [1]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Brendan McMahan, Ilya Mironov, Kunal Talwar, and Li Zhang

    Martin Abadi, Andy Chu, Ian Goodfellow, H. Brendan McMahan, Ilya Mironov, Kunal Talwar, and Li Zhang. 2016. https://doi.org/10.1145/2976749.2978318 Deep learning with differential privacy . In Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, CCS '16, page 308–318, New York, NY, USA. Association for Computing Machinery

    arXiv 2016

  4. [4]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774

    arXiv 2023

  5. [6]

    Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel Ziegler, Jeffrey Wu, Clemens Winter, Chris Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gr...

    2020

  6. [7]

    Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin...

    arXiv 2020

  7. [8]

    Kuntai Cai, Xiaoyu Lei, Jianxin Wei, and Xiaokui Xiao. 2021. https://doi.org/10.14778/3476249.3476272 Data synthesis via differentially private markov random fields . Proc. VLDB Endow., 14(11):2190–2202

    arXiv 2021

  8. [9]

    Lynn Chua, Badih Ghazi, Yangsibo Huang, Pritish Kamath, Ravi Kumar, Daogao Liu, Pasin Manurangsi, Amer Sinha, and Chiyuan Zhang. 2024. https://openreview.net/forum?id=Jd0bCD12DS Mind the privacy unit! user-level differential privacy for language model fine-tuning . In First Conference on Language Modeling

    work page 2024

Show all 41 references
  1. [10]

    Junyoung Chung, Caglar Gulcehre, KyungHyun Cho, and Yoshua Bengio. 2014. Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555

    2014 arXiv

  2. [11]

    Cynthia Dwork, Aaron Roth, et al. 2014. The algorithmic foundations of differential privacy. Foundations and Trends in Theoretical Computer Science , 9(3--4):211--407

    2014

  3. [12]

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. Advances in neural information processing systems, 27

    2014

  4. [13]

    Moritz Hardt, Katrina Ligett, and Frank Mcsherry. 2012. https://proceedings.neurips.cc/paper_files/paper/2012/file/208e43f0e45c4c78cafadb83d2888cb6-Paper.pdf A simple and practical algorithm for differentially private data release . In Advances in Neural Information Processing...

    2012

  5. [14]

    Hearst, S.T

    M.A. Hearst, S.T. Dumais, E. Osuna, J. Platt, and B. Scholkopf. 1998. https://doi.org/10.1109/5254.708428 Support vector machines . IEEE Intelligent Systems and their Applications, 13(4):18--28

    1998

  6. [15]

    Sepp Hochreiter and J\" u rgen Schmidhuber. 1997. https://doi.org/10.1162/neco.1997.9.8.1735 Long short-term memory . Neural Comput., 9(8):1735–1780

    1997 doi

  7. [16]

    Wang, Chenhui Zhang, Zhangheng LI, Bo Li, and Zhangyang Wang

    Junyuan Hong, Jiachen T. Wang, Chenhui Zhang, Zhangheng LI, Bo Li, and Zhangyang Wang. 2024. https://openreview.net/forum?id=Ifz3IgsEPX DP - OPT : Make large language model your privacy-preserving prompt engineer . In The Twelfth International Conference on Learning Representations

    2024

  8. [17]

    Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, et al. 2024. Gpt-4o system card. arXiv preprint arXiv:2410.21276

    2024 arXiv

  9. [18]

    Ismail Keshta and Ammar Odeh. 2021. Security and privacy of electronic health records: Concerns and challenges. Egyptian Informatics Journal, 22(2):177--183

    2021

  10. [19]

    Kibriya, Eibe Frank, Bernhard Pfahringer, and Geoffrey Holmes

    Ashraf M. Kibriya, Eibe Frank, Bernhard Pfahringer, and Geoffrey Holmes. 2004. https://doi.org/10.1007/978-3-540-30549-1_43 Multinomial naive bayes for text categorization revisited . In Proceedings of the 17th Australian Joint Conference on Advances in Artificial Intelligence...

    2004 doi

  11. [20]

    Diederik P Kingma, Max Welling, et al. 2019. An introduction to variational autoencoders. Foundations and Trends in Machine Learning , 12(4):307--392

    2019

  12. [21]

    Ruibo Liu, Jerry Wei, Fangyu Liu, Chenglei Si, Yanzhe Zhang, Jinmeng Rao, Steven Zheng, Daiyi Peng, Diyi Yang, Denny Zhou, et al. 2024. https://arxiv.org/abs/2404.07503 Best practices and lessons learned on synthetic data for language models . arXiv e-prints, pages arXiv--2404

    2024 arXiv

  13. [22]

    Long et al

    Y. Long et al. 2021. https://papers.nips.cc/paper/2021/file/02d6c61c1442118e8b0f97e2c90f76d7-Paper.pdf G-pate: Scalable differentially private data generator via private aggregation of teacher discriminators . In Proceedings of the 35th Conference on Neural Information Process...

    2021

  14. [23]

    Tina Marjanov, Maria Konstantinou, Magdalena J \'o \'z wiak, and Dayana Spagnuelo. 2023. Data security on the ground: Investigating technical and legal requirements under the gdpr. Proceedings on Privacy Enhancing Technologies

    2023

  15. [24]

    Md Mahadi Hasan Nahid. 2024. https://doi.org/10.7939/r3-ckmh-a783 Improving table reasoning through table decomposition and normalization . Univerity of Alberta

    2024 doi

  16. [25]

    Md Mahadi Hasan Nahid and Davood Rafiei. 2024 a . https://doi.org/10.18653/v1/2024.findings-emnlp.203 N orm T ab: Improving symbolic reasoning in LLM s through tabular data normalization . In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 3569--35...

    2024 doi

  17. [26]

    Md Mahadi Hasan Nahid and Davood Rafiei. 2024 b . https://doi.org/10.18653/v1/2024.naacl-long.320 T ab SQL ify: Enhancing reasoning capabilities of LLM s through table decomposition . In Proceedings of the 2024 Conference of the North American Chapter of the Association for Co...

    2024 doi

  18. [27]

    Mir Tafseer Nayeem and Davood Rafiei. 2024. https://doi.org/10.18653/v1/2024.emnlp-main.277 K id LM : Advancing language models for children -- early insights and future directions . In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pag...

    2024 doi

  19. [28]

    Midas Nouwens, Ilaria Liccardi, Michael Veale, David Karger, and Lalana Kagal. 2020. Dark patterns after the gdpr: Scraping consent pop-ups and demonstrating their influence. In Proceedings of the 2020 CHI conference on human factors in computing systems, pages 1--13

    2020

  20. [29]

    Jeffrey Pennington, Richard Socher, and Christopher Manning. 2014. https://doi.org/10.3115/v1/D14-1162 G lo V e: Global vectors for word representation . In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing ( EMNLP ) , pages 1532--1543, Doh...

    2014 doi

  21. [30]

    Lucas Rosenblatt, Bernease Herman, Anastasia Holovenko, Wonkwon Lee, Joshua Loftus, Elizabeth McKinnie, Taras Rumezhak, Andrii Stadnik, Bill Howe, and Julia Stoyanovich. 2023. https://doi.org/10.14778/3611479.3611517 Epistemic parity: Reproducibility as an evaluation metric fo...

    2023

  22. [31]

    Lucas Rosenblatt, Xiaoyan Liu, Samira Pouyanfar, Eduardo de Leon, Anuj Desai, and Joshua Allen. 2020. https://arxiv.org/abs/2011.05537 Differentially private synthetic data: Applied evaluations and enhancements . Preprint, arXiv:2011.05537

    2020 arXiv

  23. [32]

    Nikita Samarin, Shayna Kothari, Zaina Siyed, Oscar Bjorkman, Reena Yuan, Primal Wijesekera, Noura Alomar, Jordan Fischer, Chris Hoofnagle, and Serge Egelman. 2023. Lessons in vcr repair: Compliance of android app developers with the california consumer privacy act (ccpa). Proc...

    2023

  24. [33]

    Reza Shokri, Marco Stronati, Congzheng Song, and Vitaly Shmatikov. 2017. Membership inference attacks against machine learning models. In 2017 IEEE symposium on security and privacy (SP), pages 3--18. IEEE

    2017

  25. [34]

    Theresa Stadler, Bristena Oprisanu, and Carmela Troncoso. 2022. https://www.usenix.org/conference/usenixsecurity22/presentation/stadler Synthetic data – anonymisation groundhog day . In 31st USENIX Security Symposium (USENIX Security 22), pages 1451--1468, Boston, MA. USENIX A...

    2022

  26. [35]

    Xinyu Tang, Richard Shin, Huseyin A Inan, Andre Manoel, Fatemehsadat Mireshghallah, Zinan Lin, Sivakanth Gopi, Janardhan Kulkarni, and Robert Sim. 2024. https://openreview.net/forum?id=oZtt0pRnOl Privacy-preserving in-context learning with differentially private few-shot gener...

    2024

  27. [36]

    Gemini Team, Petko Georgiev, Ving Ian Lei, Ryan Burnell, Libin Bai, Anmol Gulati, Garrett Tanzer, Damien Vincent, Zhufeng Pan, Shibo Wang, et al. 2024. Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. arXiv preprint arXiv:2403.05530

    2024 arXiv

  28. [37]

    Reihaneh Torkzadehmahani, Peter Kairouz, and Benedict Paten. 2019. Dp-cgan: Differentially private synthetic data and label generation. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 98--104. IEEE Computer Society

    2019

  29. [38]

    Giuseppe Vietri, Grace Tian, Mark Bun, Thomas Steinke, and Steven Wu. 2020. https://proceedings.mlr.press/v119/vietri20b.html New oracle-efficient algorithms for private synthetic data release . In Proceedings of the 37th International Conference on Machine Learning, volume 11...

    2020

  30. [39]

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824--24837

    2022

  31. [40]

    Zhang, G

    J. Zhang, G. Cormode, C.M. Procopiuc, D. Srivastava, and X. Xiao. 2017. https://doi.org/10.1145/3035918.3035919 PrivBayes : Private data release via bayesian networks . In Proceedings of the 2017 ACM SIGMOD International Conference on Management of Data, pages 1423--1434, New ...

    2017

  32. [41]

    Xiang Zhang, Junbo Zhao, and Yann LeCun. 2015. Character-level convolutional networks for text classification. Advances in neural information processing systems, 28

    2015

  33. [42]

    Zhikun Zhang, Tianhao Wang, Ninghui Li, Jean Honorio, Michael Backes, Shibo He, Jiming Chen, and Yang Zhang. 2021. https://www.usenix.org/conference/usenixsecurity21/presentation/zhang-zhikun PrivSyn : Differentially private data synthesis . In 30th USENIX Security Symposium (...

    2021

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.