---
title: "Private Seeds, Public LLMs — paper by Ma and Rajtmajer (2026)"
description: "The authors propose RPSG, a three-stage pipeline that starts from actual private samples: Each seed is PII-masked with… 66 references."
authors: ["Qian Ma", "Sarah Rajtmajer"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.10/"
doi: "10.18653/v1/2026.findings-acl.10"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/ma-2026-private-seeds-public-llms-realistic/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/ma-2026-private-seeds-public-llms-realistic/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/45/) / Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation

# Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation

**Authors:** [Qian Ma](https://md.research.elamur.ai/free/en/authors/qian-ma/) and Sarah Rajtmajer · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.10/) · [PDF](https://aclanthology.org/2026.findings-acl.10.pdf) · [DOI 10.18653/v1/2026.findings-acl.10](https://doi.org/10.18653/v1/2026.findings-acl.10) · [OpenAlex](https://openalex.org/W7166886659)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Sensitive text, such as social media posts with self-disclosures or clinical notes, often cannot be shared. Synthetic stand-ins must therefore stay useful without exposing anyone. DP fine-tuning is expensive and does not work for models available only through an API. Prompt-only generation tends to drift away from the real data. The authors propose RPSG, a three-stage pipeline that starts from actual private samples:

1. Each seed is PII-masked with regex rules. An abstraction model then turns it into several sentiment-consistent abstract versions, and one is chosen through a Gaussian differential privacy mechanism.
2. An LLM produces variations of the chosen candidate.
3. The outputs are filtered by cosine similarity and NLL to curb memorization, and leftover PII is redacted.

RPSG is tested on a PubMed benchmark and a newly collected Reddit corpus. It is compared with DP-SGD, AUG-PE and RUPTA across several privacy budgets and LLMs. It shows competitive or better utility, diversity and efficiency, and membership inference attacks perform close to chance. This makes it a practical route to shareable replicas of private corpora.

## Key points
- **Baselines:** RPSG is compared with three kinds of methods:
  - DP-SGD, which is gradient-based;
  - AUG-PE, which is prompt-based;
  - RUPTA, which rewrites each text one-to-one.
- **Privacy results:** Membership inference attacks reach an AUC of about 50%, roughly random guessing. PII extraction rates are 0.5–1.2% of samples; newer models like GPT-4o-mini leak less than GPT-2.
- **Efficiency:** RPSG runs 1.22x–1.38x faster than the baselines in GPU hours.
- **Abstraction fidelity:** Sentiment matches between seeds and their abstractions in 86.1%, 89.6% and 86.5% of cases (subsets of 100, 200 and 500 samples). Abstractions are capped at 150 tokens.
- **Limitations and availability:**
  - Filtering thresholds must be tuned for each model.
  - Corpus-level re-profiling, where an attacker combines many samples, was not evaluated.
  - Code is at github.com/masonmq/rpsg; the Reddit data is available on request.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Pennsylvania State University](https://md.research.elamur.ai/free/en/universities/pennsylvania-state-university/) — University, State College, United States · 67 works · 5,865 citations

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works

🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works

## Cite

Cite as: Qian Ma and Sarah Rajtmajer. 2026. Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{ma2026private,
  title = {{Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation}},
  author = {Qian Ma and Sarah Rajtmajer},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.10},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.10}
}
```

- [← previous Privacy VITA: a new multilingual and multimodal annotated video corpus to evaluate anonymization systems](https://md.research.elamur.ai/free/en/papers/rico-2026-privacy-vita-new-multilingual-multimodal/)
- [next → priyam_saha17 at SVELA: A Feature-Centric Pipeline for Verifying Selective Forgetting in Large Language Models](https://md.research.elamur.ai/free/en/papers/saha-2026-priyam-saha17-svela-feature-centric/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/45/)

## References

66 references, 21 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Martín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan, Ilya Mironov, Kunal Talwar, and Li Zhang. 2016. Deep learning with differential privacy. In Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, Vienna, Austria, October 24-28, 2016, pages 308–318. ACM. [find](https://scholar.google.com/scholar?q=Mart%C3%ADn+Abadi%2C+Andy+Chu%2C+Ian+J%2E+Goodfellow%2C+H%2E+Brendan+McMahan%2C+Ilya+Mironov%2C+Kunal+Talwar%2C+and+Li+Zhang%2E+2016%2E+Deep+learning+with+differential+privacy%2E+In+Proceedings+of+the+2016+ACM+SIGSAC+Conference)
2. Marah I Abdin, Jyoti Aneja, Harkirat S. Behl, Sébastien Bubeck, Ronen Eldan, Suriya Gunasekar, Michael Harrison, Russell J. Hewett, Mojan Javaheripi, Piero Kauffmann, James R. Lee, Yin Tat Lee, Yuanzhi Li, Weishung Liu, Caio C. T. Mendes, Anh Nguyen, Eric Price, Gustavo de Rosa, Olli Saarikivi, and… — [Phi-4 Technical Report](https://md.research.elamur.ai/free/en/papers/abdin-2024-phi-4-technical-report/)
3. Atilla Akkus, Mingjie Li, Junjie Chu, Michael Backes, Yang Zhang, and Sinem Sav. 2025. Generated data with fake privacy: Hidden dangers of fine-tuning large language models on generated data. CoRR, abs/2409.11423. — [arXiv:2409.11423](https://arxiv.org/abs/2409.11423) [find](https://scholar.google.com/scholar?q=Atilla+Akkus%2C+Mingjie+Li%2C+Junjie+Chu%2C+Michael+Backes%2C+Yang+Zhang%2C+and+Sinem+Sav%2E+2025%2E+Generated+data+with+fake+privacy%3A+Hidden+dangers+of+fine%2Dtuning+large+language+models+on+generated+data%2E+CoRR%2C+ab)
4. Syeda Nahida Akter, Shrimai Prabhumoye, John Kamalu, Sanjeev Satheesh, Eric Nyberg, Mostofa Patwary, Mohammad Shoeybi, and Bryan Catanzaro. 2024. MIND: math informed synthetic dialogues for pretraining llms. CoRR, abs/2410.12881. [find](https://scholar.google.com/scholar?q=Syeda+Nahida+Akter%2C+Shrimai+Prabhumoye%2C+John+Kamalu%2C+Sanjeev+Satheesh%2C+Eric+Nyberg%2C+Mostofa+Patwary%2C+Mohammad+Shoeybi%2C+and+Bryan+Catanzaro%2E+2024%2E+MIND%3A+math+informed+synthetic+dialogues+for+pretrainin)
5. Loubna Ben Allal, Raymond Li, Denis Kocetkov, Chenghao Mou, Christopher Akiki, Carlos Muñoz Ferrandis, Niklas Muennighoff, Mayank Mishra, Alex Gu, Manan Dey, Logesh Kumar Umapathi, Carolyn Jane Anderson, Yangtian Zi, Joel Lamy-Poirier, Hailey Schoelkopf, Sergey Troshin, Dmitry Abulkhanov, Manuel Rom… — [SantaCoder: don't reach for the stars!](https://md.research.elamur.ai/free/en/papers/allal-2023-santacoder-don-t-reach-stars/)
6. Anthropic. 2025. Claude 3.7 sonnet: Hybrid reasoning model. [https://www.anthropic.com/news/](https://www.anthropic.com/news/) introducing-citations-api. [↗](https://www.anthropic.com/news/)
7. Tamar Ashuri and Ruth Halperin. 2024. Online self-disclosure: An interdisciplinary literature review of 10 years of research. New Media & Society, page 14614448241247313. [find](https://scholar.google.com/scholar?q=Tamar+Ashuri+and+Ruth+Halperin%2E+2024%2E+Online+self%2Ddisclosure%3A+An+interdisciplinary+literature+review+of+10+years+of+research%2E+New+Media+%26+Society%2C+page+14614448241247313%2E)
8. Yushi Bai, Jiajie Zhang, Xin Lv, Linzhi Zheng, Siqi Zhu, Lei Hou, Yuxiao Dong, Jie Tang, and Juanzi Li. 2024. [Longwriter: Unleashing 10,000+ word generation from long context llms](https://md.research.elamur.ai/free/en/papers/bai-2024-longwriter-unleashing-10-000/). CoRR, abs/2408.07055.
9. Raef Bassily, Adam D. Smith, and Abhradeep Thakurta. 2014. Private empirical risk minimization: Efficient algorithms and tight error bounds. In 55th IEEE Annual Symposium on Foundations of Computer Science, FOCS 2014, Philadelphia, PA, USA, October 18-21, 2014, pages 464–473. IEEE Computer Society. [find](https://scholar.google.com/scholar?q=Raef+Bassily%2C+Adam+D%2E+Smith%2C+and+Abhradeep+Thakurta%2E+2014%2E+Private+empirical+risk+minimization%3A+Efficient+algorithms+and+tight+error+bounds%2E+In+55th+IEEE+Annual+Symposium+on+Foundations+of+Computer+Sc)
10. Rishi Bommasani, Steven Wu, and Xanda Schofield. 2019. Towards private synthetic text generation. In NeurIPS 2019 Machine Learning with Guarantees Workshop. [find](https://scholar.google.com/scholar?q=Rishi+Bommasani%2C+Steven+Wu%2C+and+Xanda+Schofield%2E+2019%2E+Towards+private+synthetic+text+generation%2E+In+NeurIPS+2019+Machine+Learning+with+Guarantees+Workshop%2E)
11. Yihan Cao, Siyu Li, Yixin Liu, Zhiling Yan, Yutong Dai, Philip S. Yu, and Lichao Sun. 2023. A comprehensive survey of ai-generated content (AIGC): A history of generative AI from GAN to chatgpt. CoRR, abs/2303.04226. [find](https://scholar.google.com/scholar?q=Yihan+Cao%2C+Siyu+Li%2C+Yixin+Liu%2C+Zhiling+Yan%2C+Yutong+Dai%2C+Philip+S%2E+Yu%2C+and+Lichao+Sun%2E+2023%2E+A+comprehensive+survey+of+ai%2Dgenerated+content+%28AIGC%29%3A+A+history+of+generative+AI+from+GAN+to+chatgpt%2E+CoRR%2C)
12. Nicholas Carlini, Steve Chien, Milad Nasr, Shuang Song, Andreas Terzis, and Florian Tramèr. 2022. [Membership inference attacks from first principles](https://md.research.elamur.ai/free/en/papers/carlini-2021-membership-inference-attacks-first-principles/). In 43rd IEEE Symposium on Security and Privacy, SP 2022, San Francisco, CA, USA, May 22-26, 2022, pages 1897–1914. IEEE.
13. Nicholas Carlini, Florian Tramèr, Eric Wallace, Matthew Jagielski, Ariel Herbert-Voss, Katherine Lee, Adam Roberts, Tom B. Brown, Dawn Song, Úlfar Erlingsson, Alina Oprea, and Colin Raffel. 2021. [Extracting training data from large language models](https://md.research.elamur.ai/free/en/papers/carlini-2020-extracting-training-data-large-language/). In 30th USENIX Security Symposium, USENIX Security…
14. Munmun De Choudhury and Sushovan De. 2014. Mental health discourse on reddit: Self-disclosure, social support, and anonymity. In Proceedings of the Eighth International Conference on Weblogs and Social Media, ICWSM 2014, Ann Arbor, Michigan, USA, June 1-4, 2014. The AAAI Press. [find](https://scholar.google.com/scholar?q=Munmun+De+Choudhury+and+Sushovan+De%2E+2014%2E+Mental+health+discourse+on+reddit%3A+Self%2Ddisclosure%2C+social+support%2C+and+anonymity%2E+In+Proceedings+of+the+Eighth+International+Conference+on+Weblogs+and+Socia)
15. DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, and 81 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in llm…](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/)
16. Giulia DeSalvo, Jean-François Kagy, Lazaros Karydas, Afshin Rostamizadeh, and Sanjiv Kumar. 2024. No more hard prompts: Softsrv prompting for synthetic data generation. CoRR, abs/2410.16534. — [arXiv:2410.16534](https://arxiv.org/abs/2410.16534) [find](https://scholar.google.com/scholar?q=Giulia+DeSalvo%2C+Jean%2DFran%C3%A7ois+Kagy%2C+Lazaros+Karydas%2C+Afshin+Rostamizadeh%2C+and+Sanjiv+Kumar%2E+2024%2E+No+more+hard+prompts%3A+Softsrv+prompting+for+synthetic+data+generation%2E+CoRR%2C+abs%2F2410%2E16534%2E+%E2%80%94+%5BarXiv%3A)
17. Yao Dou, Isadora Krsek, Tarek Naous, Anubha Kabra, Sauvik Das, Alan Ritter, and Wei Xu. 2024. Reducing privacy risks in online self-disclosures with language models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Bangkok,… [find](https://scholar.google.com/scholar?q=Yao+Dou%2C+Isadora+Krsek%2C+Tarek+Naous%2C+Anubha+Kabra%2C+Sauvik+Das%2C+Alan+Ritter%2C+and+Wei+Xu%2E+2024%2E+Reducing+privacy+risks+in+online+self%2Ddisclosures+with+language+models%2E+In+Proceedings+of+the+62nd+Annual+)
18. Tingting Du, Jiyoon Kim, Anna Squicciarini, and Sarah Rajtmajer. 2024. Toward context-aware privacy enhancing technologies for online self-disclosure. In Proceedings of the AAAI Conference on Human Computation and Crowdsourcing, volume 12, pages 44– 54. [find](https://scholar.google.com/scholar?q=Tingting+Du%2C+Jiyoon+Kim%2C+Anna+Squicciarini%2C+and+Sarah+Rajtmajer%2E+2024%2E+Toward+context%2Daware+privacy+enhancing+technologies+for+online+self%2Ddisclosure%2E+In+Proceedings+of+the+AAAI+Conference+on+Human+Co)
19. Michael Duan, Anshuman Suri, Niloofar Mireshghallah, Sewon Min, Weijia Shi, Luke Zettlemoyer, Yulia Tsvetkov, Yejin Choi, David Evans, and Hannaneh Hajishirzi. 2024. [Do membership inference attacks work on large language models?](https://md.research.elamur.ai/free/en/papers/duan-2024-membership-inference-attacks-work-large/) CoRR, abs/2402.07841.
20. Cynthia Dwork and Aaron Roth. 2014. The algorithmic foundations of differential privacy. Found. Trends Theor. Comput. Sci., 9(3-4):211–407. [find](https://scholar.google.com/scholar?q=Cynthia+Dwork+and+Aaron+Roth%2E+2014%2E+The+algorithmic+foundations+of+differential+privacy%2E+Found%2E+Trends+Theor%2E+Comput%2E+Sci%2E%2C+9%283%2D4%29%3A211%E2%80%93407%2E)
21. Kennedy Edemacu and Xintao Wu. 2024. Privacy preserving prompt engineering: A survey. CoRR, abs/2404.06001. — [arXiv:2404.06001](https://arxiv.org/abs/2404.06001) [find](https://scholar.google.com/scholar?q=Kennedy+Edemacu+and+Xintao+Wu%2E+2024%2E+Privacy+preserving+prompt+engineering%3A+A+survey%2E+CoRR%2C+abs%2F2404%2E06001%2E+%E2%80%94+%5BarXiv%3A2404%2E06001%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2404%2E06001%29)
22. Ahmed Frikha, Nassim Walha, Krishna Kanth Nakka, Ricardo Mendes, Xue Jiang, and Xuebing Zhou. 2024. [Incognitext: Privacy-enhancing conditional text anonymization via llm-based private attribute randomization](https://md.research.elamur.ai/free/en/papers/frikha-2025-incognitext-privacy-enhancing-conditional-text/). CoRR, abs/2407.02956.
23. Anatoliy A. Gruzd and Ángel Hernández-García. 2018. Privacy concerns and self-disclosure in private and public uses of social media. Cyberpsychology Behav. Soc. Netw., 21(7):418–428. [find](https://scholar.google.com/scholar?q=Anatoliy+A%2E+Gruzd+and+%C3%81ngel+Hern%C3%A1ndez%2DGarc%C3%ADa%2E+2018%2E+Privacy+concerns+and+self%2Ddisclosure+in+private+and+public+uses+of+social+media%2E+Cyberpsychology+Behav%2E+Soc%2E+Netw%2E%2C+21%287%29%3A418%E2%80%93428%2E)
24. Yu Gu, Robert Tinn, Hao Cheng, Michael Lucas, Naoto Usuyama, Xiaodong Liu, Tristan Naumann, Jianfeng Gao, and Hoifung Poon. 2022. [Domain-specific language model pretraining for biomedical natural language processing](https://md.research.elamur.ai/free/en/papers/gu-2020-domain-specific-language-model-pretraining/). ACM Trans. Comput. Heal., 3(1):2:1–2:23.
25. Perttu Hämäläinen, Mikke Tavast, and Anton Kunnari. 2023. Evaluating large language models in generating synthetic HCI research data: a case study. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, CHI 2023, Hamburg, Germany, April 23-28, 2023, pages 433:1–433:19. ACM. [find](https://scholar.google.com/scholar?q=Perttu+H%C3%A4m%C3%A4l%C3%A4inen%2C+Mikke+Tavast%2C+and+Anton+Kunnari%2E+2023%2E+Evaluating+large+language+models+in+generating+synthetic+HCI+research+data%3A+a+case+study%2E+In+Proceedings+of+the+2023+CHI+Conference+on+Human+F)
26. Jochen Hartmann, Mark Heitmann, Christian Siebert, and Christina Schamp. 2023. More than a feeling: Accuracy and application of sentiment analysis. International Journal of Research in Marketing, 40(1):75–87. [find](https://scholar.google.com/scholar?q=Jochen+Hartmann%2C+Mark+Heitmann%2C+Christian+Siebert%2C+and+Christina+Schamp%2E+2023%2E+More+than+a+feeling%3A+Accuracy+and+application+of+sentiment+analysis%2E+International+Journal+of+Research+in+Marketing%2C+40%281)
27. Tianxing He, Jingyu Zhang, Tianle Wang, Sachin Kumar, Kyunghyun Cho, James R. Glass, and Yulia Tsvetkov. 2023. On the blind spots of model-based evaluation metrics for text generation. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers),… [find](https://scholar.google.com/scholar?q=Tianxing+He%2C+Jingyu+Zhang%2C+Tianle+Wang%2C+Sachin+Kumar%2C+Kyunghyun+Cho%2C+James+R%2E+Glass%2C+and+Yulia+Tsvetkov%2E+2023%2E+On+the+blind+spots+of+model%2Dbased+evaluation+metrics+for+text+generation%2E+In+Proceedings+)
28. Ari Holtzman, Jan Buys, Li Du, Maxwell Forbes, and Yejin Choi. 2020. [The curious case of neural text degeneration](https://md.research.elamur.ai/free/en/papers/holtzman-2019-curious-case-neural-text-degeneration/). In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net.
29. Charlie Hou, Akshat Shrivastava, Hongyuan Zhan, Rylan Conway, Trang Le, Adithya Sagar, Giulia Fanti, and Daniel Lazar. 2024. Pre-text: Training language models on private federated data in the age of llms. In Forty-first International Conference on Machine Learning, ICML 2024, Vienna, Austria, July… [find](https://scholar.google.com/scholar?q=Charlie+Hou%2C+Akshat+Shrivastava%2C+Hongyuan+Zhan%2C+Rylan+Conway%2C+Trang+Le%2C+Adithya+Sagar%2C+Giulia+Fanti%2C+and+Daniel+Lazar%2E+2024%2E+Pre%2Dtext%3A+Training+language+models+on+private+federated+data+in+the+age+of+)
30. Nicholas I-Hsien Kuo, Blanca Gallego, and Louisa Jorm. 2024. Masked clinical modelling: A framework for synthetic and augmented survival data generation. CoRR, abs/2410.16811. — [arXiv:2410.16811](https://arxiv.org/abs/2410.16811) [find](https://scholar.google.com/scholar?q=Nicholas+I%2DHsien+Kuo%2C+Blanca+Gallego%2C+and+Louisa+Jorm%2E+2024%2E+Masked+clinical+modelling%3A+A+framework+for+synthetic+and+augmented+survival+data+generation%2E+CoRR%2C+abs%2F2410%2E16811%2E+%E2%80%94+%5BarXiv%3A2410%2E16811%5D%28htt)
31. Alexey Kurakin, Natalia Ponomareva, Umar Syed, Liam MacDermed, and Andreas Terzis. 2023a. Harnessing large-language models to generate private synthetic text. arXiv preprint arXiv:2306.01684. attacks against language models via neighbourhood comparison. In Findings of the Association for Computation… — [arXiv:2306.01684](https://arxiv.org/abs/2306.01684) [find](https://scholar.google.com/scholar?q=Alexey+Kurakin%2C+Natalia+Ponomareva%2C+Umar+Syed%2C+Liam+MacDermed%2C+and+Andreas+Terzis%2E+2023a%2E+Harnessing+large%2Dlanguage+models+to+generate+private+synthetic+text%2E+arXiv+preprint+arXiv%3A2306%2E01684%2E+attacks+)
32. Meta. 2024. Llama 3.3: Multilingual large language model. [https://huggingface.co/meta-llama/](https://huggingface.co/meta-llama/) Llama-3.3-70B-Instruct. [↗](https://huggingface.co/meta-llama/)
33. Alexey Kurakin, Natalia Ponomareva, Umar Syed, Liam MacDermed, and Andreas Terzis. 2023b. Harnessing large-language models to generate private synthetic text. CoRR, abs/2306.01684. — [arXiv:2306.01684](https://arxiv.org/abs/2306.01684) [find](https://scholar.google.com/scholar?q=Alexey+Kurakin%2C+Natalia+Ponomareva%2C+Umar+Syed%2C+Liam+MacDermed%2C+and+Andreas+Terzis%2E+2023b%2E+Harnessing+large%2Dlanguage+models+to+generate+private+synthetic+text%2E+CoRR%2C+abs%2F2306%2E01684%2E+%E2%80%94+%5BarXiv%3A2306%2E01684)
34. Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Veselin Stoyanov, and Luke Zettlemoyer. 2019. [BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension](https://md.research.elamur.ai/free/en/papers/lewis-2020-bart-denoising-sequence-sequence-pre/). CoRR, abs/1910.13461.
35. Fatemehsadat Mireshghallah, Archit Uniyal, Tianhao Wang, David Evans, and Taylor Berg-Kirkpatrick. 2022. Memorization in NLP fine-tuning methods. CoRR, abs/2205.12506. — [arXiv:2205.12506](https://arxiv.org/abs/2205.12506) [find](https://scholar.google.com/scholar?q=Fatemehsadat+Mireshghallah%2C+Archit+Uniyal%2C+Tianhao+Wang%2C+David+Evans%2C+and+Taylor+Berg%2DKirkpatrick%2E+2022%2E+Memorization+in+NLP+fine%2Dtuning+methods%2E+CoRR%2C+abs%2F2205%2E12506%2E+%E2%80%94+%5BarXiv%3A2205%2E12506%5D%28https%3A%2F%2Farx)
36. Ilya Mironov. 2017. Rényi differential privacy. In 30th IEEE Computer Security Foundations Symposium, CSF 2017, Santa Barbara, CA, USA, August 21-25, 2017, pages 263–275. IEEE Computer Society. [find](https://scholar.google.com/scholar?q=Ilya+Mironov%2E+2017%2E+R%C3%A9nyi+differential+privacy%2E+In+30th+IEEE+Computer+Security+Foundations+Symposium%2C+CSF+2017%2C+Santa+Barbara%2C+CA%2C+USA%2C+August+21%2D25%2C+2017%2C+pages+263%E2%80%93275%2E+IEEE+Computer+Society%2E)
37. Qinbin Li, Junyuan Hong, Chulin Xie, Jeffrey Tan, Rachel Xin, Junyi Hou, Xavier Yin, Zhun Wang, Dan Hendrycks, Zhangyang Wang, Bo Li, Bingsheng He, and Dawn Song. 2024a. LLM-PBE: assessing data privacy in large language models. Proc. VLDB Endow., 17(11):3201–3214. [find](https://scholar.google.com/scholar?q=Qinbin+Li%2C+Junyuan+Hong%2C+Chulin+Xie%2C+Jeffrey+Tan%2C+Rachel+Xin%2C+Junyi+Hou%2C+Xavier+Yin%2C+Zhun+Wang%2C+Dan+Hendrycks%2C+Zhangyang+Wang%2C+Bo+Li%2C+Bingsheng+He%2C+and+Dawn+Song%2E+2024a%2E+LLM%2DPBE%3A+assessing+data+privac)
38. Ehsan Montahaei, Danial Alihosseini, and Mahdieh Soleymani Baghshah. 2019. Jointly measuring diversity and quality in text generation models. CoRR, abs/1904.03971. [find](https://scholar.google.com/scholar?q=Ehsan+Montahaei%2C+Danial+Alihosseini%2C+and+Mahdieh+Soleymani+Baghshah%2E+2019%2E+Jointly+measuring+diversity+and+quality+in+text+generation+models%2E+CoRR%2C+abs%2F1904%2E03971%2E)
39. Xiaochuan Li, Zichun Yu, and Chenyan Xiong. 2024b. Montessori-instruct: Generate influential training data tailored for student learning. CoRR, abs/2410.14208. — [arXiv:2410.14208](https://arxiv.org/abs/2410.14208) [find](https://scholar.google.com/scholar?q=Xiaochuan+Li%2C+Zichun+Yu%2C+and+Chenyan+Xiong%2E+2024b%2E+Montessori%2Dinstruct%3A+Generate+influential+training+data+tailored+for+student+learning%2E+CoRR%2C+abs%2F2410%2E14208%2E+%E2%80%94+%5BarXiv%3A2410%2E14208%5D%28https%3A%2F%2Farxiv%2Eorg%2Fa)
40. OpenAI. 2025. Introducing gpt-5. [https://openai.com/index/introducing-gpt-5/](https://openai.com/index/introducing-gpt-5/). [↗](https://openai.com/index/introducing-gpt-5/)
41. Lin Long, Rui Wang, Ruixuan Xiao, Junbo Zhao, Xiao Ding, Gang Chen, and Haobo Wang. 2024. [On llms-driven synthetic data generation, curation, and evaluation: A survey](https://md.research.elamur.ai/free/en/papers/long-2024-llms-driven-synthetic-data-generation/). In Findings of the Association for Computational Linguistics, ACL 2024, Bangkok, Thailand and virtual meeting, August 11-16, 2024, p…
42. Yingzhou Lu, Huazheng Wang, and Wenqi Wei. 2023. Machine learning for synthetic data generation: a review. CoRR, abs/2302.04062. — [arXiv:2302.04062](https://arxiv.org/abs/2302.04062) [find](https://scholar.google.com/scholar?q=Yingzhou+Lu%2C+Huazheng+Wang%2C+and+Wenqi+Wei%2E+2023%2E+Machine+learning+for+synthetic+data+generation%3A+a+review%2E+CoRR%2C+abs%2F2302%2E04062%2E+%E2%80%94+%5BarXiv%3A2302%2E04062%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2302%2E04062%29)
43. Sadhika Malladi, Tianyu Gao, Eshaan Nichani, Alex Damian, Jason D. Lee, Danqi Chen, and Sanjeev Arora. 2023. [Fine-tuning language models with just forward passes](https://md.research.elamur.ai/free/en/papers/malladi-2023-fine-tuning-language-models-just/). In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, N…
44. Justus Mattern, Zhijing Jin, Benjamin Weggenmann, Bernhard Schölkopf, and Mrinmaya Sachan. 2022. Differentially private language models for secure data sharing. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022, Abu Dhabi, United Arab Emirates, Dec… [find](https://scholar.google.com/scholar?q=Justus+Mattern%2C+Zhijing+Jin%2C+Benjamin+Weggenmann%2C+Bernhard+Sch%C3%B6lkopf%2C+and+Mrinmaya+Sachan%2E+2022%2E+Differentially+private+language+models+for+secure+data+sharing%2E+In+Proceedings+of+the+2022+Conference+o)
45. Justus Mattern, Fatemehsadat Mireshghallah, Zhijing Jin, Bernhard Schölkopf, Mrinmaya Sachan, and Taylor Berg-Kirkpatrick. 2023. Membership inference OpenAI. 2023. [GPT-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). abs/2303.08774.
46. CoRR, Nicolas Papernot, Martín Abadi, Úlfar Erlingsson, Ian J. Goodfellow, and Kunal Talwar. 2017. Semi-supervised knowledge transfer for deep learning from private training data. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Tr… [find](https://scholar.google.com/scholar?q=CoRR%2C+Nicolas+Papernot%2C+Mart%C3%ADn+Abadi%2C+%C3%9Alfar+Erlingsson%2C+Ian+J%2E+Goodfellow%2C+and+Kunal+Talwar%2E+2017%2E+Semi%2Dsupervised+knowledge+transfer+for+deep+learning+from+private+training+data%2E+In+5th+International)
47. Nicolas Papernot, Shuang Song, Ilya Mironov, Ananth Raghunathan, Kunal Talwar, and Úlfar Erlingsson. 2018. Scalable private learning with PATE. In 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30 -May 3, 2018, Conference Track Proceedings. Open-Rev… [find](https://scholar.google.com/scholar?q=Nicolas+Papernot%2C+Shuang+Song%2C+Ilya+Mironov%2C+Ananth+Raghunathan%2C+Kunal+Talwar%2C+and+%C3%9Alfar+Erlingsson%2E+2018%2E+Scalable+private+learning+with+PATE%2E+In+6th+International+Conference+on+Learning+Representati)
48. Nils Reimers and Iryna Gurevych. 2019. [Sentence-bert: Sentence embeddings using siamese bert-networks](https://md.research.elamur.ai/free/en/papers/reimers-2019-sentence-bert-sentence-embeddings-siamese/). In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, EMNLP-IJCNLP 2019, Hong Kong, Ch…
49. Anna Rumshisky, Marzyeh Ghassemi, Tristan Naumann, Peter Szolovits, VM Castro, TH McCoy, and RH Perlis. 2016. Predicting early psychiatric readmission with natural language processing of narrative discharge summaries. Translational psychiatry, 6(10):e921–e921. [find](https://scholar.google.com/scholar?q=Anna+Rumshisky%2C+Marzyeh+Ghassemi%2C+Tristan+Naumann%2C+Peter+Szolovits%2C+VM+Castro%2C+TH+McCoy%2C+and+RH+Perlis%2E+2016%2E+Predicting+early+psychiatric+readmission+with+natural+language+processing+of+narrative+dis)
50. Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. 2019. [Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter](https://md.research.elamur.ai/free/en/papers/sanh-2019-distilbert-distilled-version-bert-smaller/). In NeurIPS EMC 2 W orkshop.
51. Anup Shirgaonkar, Nikhil Pandey, Nazmiye Ceren Abay, Tolga Aktas, and Vijay Aski. 2024. Knowledge distillation using frontier open-source llms: Generalizability and the role of synthetic data. CoRR, abs/2410.18588. — [arXiv:2410.18588](https://arxiv.org/abs/2410.18588) [find](https://scholar.google.com/scholar?q=Anup+Shirgaonkar%2C+Nikhil+Pandey%2C+Nazmiye+Ceren+Abay%2C+Tolga+Aktas%2C+and+Vijay+Aski%2E+2024%2E+Knowledge+distillation+using+frontier+open%2Dsource+llms%3A+Generalizability+and+the+role+of+synthetic+data%2E+CoRR%2C+a)
52. Reza Shokri, Marco Stronati, Congzheng Song, and Vitaly Shmatikov. 2017. [Membership inference attacks against machine learning models](https://md.research.elamur.ai/free/en/papers/shokri-2016-membership-inference-attacks-against-machine/). In 2017 IEEE Symposium on Security and Privacy, SP 2017, San Jose, CA, USA, May 22-26, 2017, pages 3–18. IEEE Computer Society.
53. Ruixiang Tang, Xiaotian Han, Xiaoqian Jiang, and Xia Hu. 2023. Does synthetic data generation of llms help clinical text mining? CoRR, abs/2303.04360. via foundation model apis 2: Text. In Forty-first International Conference on Machine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024. OpenRev… — [arXiv:2303.04360](https://arxiv.org/abs/2303.04360) [find](https://scholar.google.com/scholar?q=Ruixiang+Tang%2C+Xiaotian+Han%2C+Xiaoqian+Jiang%2C+and+Xia+Hu%2E+2023%2E+Does+synthetic+data+generation+of+llms+help+clinical+text+mining%3F+CoRR%2C+abs%2F2303%2E04360%2E+via+foundation+model+apis+2%3A+Text%2E+In+Forty%2Dfirst)
54. Tianyu Yang, Xiaodan Zhu, and Iryna Gurevych. 2025. Robust utility-preserving text anonymization based on large language models. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2025, Vienna, Austria, July 27 - August 1, 2025, pa… [find](https://scholar.google.com/scholar?q=Tianyu+Yang%2C+Xiaodan+Zhu%2C+and+Iryna+Gurevych%2E+2025%2E+Robust+utility%2Dpreserving+text+anonymization+based+on+large+language+models%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the+Association+for+Comput)
55. Da Yu, Arturs Backurs, Sivakanth Gopi, Huseyin Inan, Janardhan Kulkarni, Zinan Lin, Chulin Xie, Huishuai Zhang, and Wanrong Zhang. 2023. Training private and efficient language models with synthetic data from llms. In Socially Responsible Language Modelling Research. [find](https://scholar.google.com/scholar?q=Da+Yu%2C+Arturs+Backurs%2C+Sivakanth+Gopi%2C+Huseyin+Inan%2C+Janardhan+Kulkarni%2C+Zinan+Lin%2C+Chulin+Xie%2C+Huishuai+Zhang%2C+and+Wanrong+Zhang%2E+2023%2E+Training+private+and+efficient+language+models+with+synthetic+d)
56. Iulia Turc, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. [Well-read students learn better: The impact of student initialization on knowledge distillation](https://md.research.elamur.ai/free/en/papers/turc-2019-well-read-students-learn-better/). CoRR, abs/1908.08962.
57. Jason A. Walonoski, Mark Kramer, Joseph Nichols, Andre Quina, Chris Moesel, Dylan Hall, Carlton Duffett, Kudakwashe Dube, Thomas Gallagher, and Scott McLachlan. 2018. Synthea: An approach, method, and software mechanism for generating synthetic patients and the synthetic electronic health care recor… [find](https://scholar.google.com/scholar?q=Jason+A%2E+Walonoski%2C+Mark+Kramer%2C+Joseph+Nichols%2C+Andre+Quina%2C+Chris+Moesel%2C+Dylan+Hall%2C+Carlton+Duffett%2C+Kudakwashe+Dube%2C+Thomas+Gallagher%2C+and+Scott+McLachlan%2E+2018%2E+Synthea%3A+An+approach%2C+method%2C+and)
58. Boxin Wang, Weixin Chen, Hengzhi Pei, Chulin Xie, Mintong Kang, Chenhui Zhang, Chejian Xu, Zidi Xiong, Ritik Dutta, Rylan Schaeffer, Sang T. Truong, Simran Arora, Mantas Mazeika, Dan Hendrycks, Zinan Lin, Yu Cheng, Sanmi Koyejo, Dawn Song, and Bo Li. 2023. [Decodingtrust: A comprehensive assessment o…](https://md.research.elamur.ai/free/en/papers/wang-2023-decodingtrust-comprehensive-assessment-trustworthiness-gpt/)
59. Yuxin Wen, Leo Marchyok, Sanghyun Hong, Jonas Geiping, Tom Goldstein, and Nicholas Carlini. 2024. Privacy backdoors: Enhancing membership inference through poisoning pre-trained models. CoRR, abs/2404.01231. — [arXiv:2404.01231](https://arxiv.org/abs/2404.01231) [find](https://scholar.google.com/scholar?q=Yuxin+Wen%2C+Leo+Marchyok%2C+Sanghyun+Hong%2C+Jonas+Geiping%2C+Tom+Goldstein%2C+and+Nicholas+Carlini%2E+2024%2E+Privacy+backdoors%3A+Enhancing+membership+inference+through+poisoning+pre%2Dtrained+models%2E+CoRR%2C+abs%2F2404)
60. Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pier-ric Cistac, Tim Rault, Rémi Louf, Morgan Funtowicz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, and 3 others. 2020. [Transf…](https://md.research.elamur.ai/free/en/papers/wolf-2020-transformers-state-art-natural-language/)
61. Da Yu, Saurabh Naik, Arturs Backurs, Sivakanth Gopi, Huseyin A. Inan, Gautam Kamath, Janardhan Kulkarni, Yin Tat Lee, Andre Manoel, Lukas Wutschitz, Sergey Yekhanin, and Huishuai Zhang. 2022. Differentially private fine-tuning of language models. In The Tenth International Conference on Learning Rep… [find](https://scholar.google.com/scholar?q=Da+Yu%2C+Saurabh+Naik%2C+Arturs+Backurs%2C+Sivakanth+Gopi%2C+Huseyin+A%2E+Inan%2C+Gautam+Kamath%2C+Janardhan+Kulkarni%2C+Yin+Tat+Lee%2C+Andre+Manoel%2C+Lukas+Wutschitz%2C+Sergey+Yekhanin%2C+and+Huishuai+Zhang%2E+2022%2E+Differen)
62. Xiang Yue, Huseyin A. Inan, Xuechen Li, Girish Kumar, Julia McAnallen, Hoda Shajari, Huan Sun, David Levitan, and Robert Sim. 2023. Synthetic text generation with differential privacy: A simple and practical recipe. In Proceedings of the 61st Annual Meeting of the Association for Computational Lingu… [find](https://scholar.google.com/scholar?q=Xiang+Yue%2C+Huseyin+A%2E+Inan%2C+Xuechen+Li%2C+Girish+Kumar%2C+Julia+McAnallen%2C+Hoda+Shajari%2C+Huan+Sun%2C+David+Levitan%2C+and+Robert+Sim%2E+2023%2E+Synthetic+text+generation+with+differential+privacy%3A+A+simple+and+pr)
63. Hanna Yukhymenko, Robin Staab, Mark Vero, and Martin T. Vechev. 2024. A synthetic dataset for personal attribute inference. CoRR, abs/2406.07217. — [arXiv:2406.07217](https://arxiv.org/abs/2406.07217) [find](https://scholar.google.com/scholar?q=Hanna+Yukhymenko%2C+Robin+Staab%2C+Mark+Vero%2C+and+Martin+T%2E+Vechev%2E+2024%2E+A+synthetic+dataset+for+personal+attribute+inference%2E+CoRR%2C+abs%2F2406%2E07217%2E+%E2%80%94+%5BarXiv%3A2406%2E07217%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2406%2E07217%29)
64. Yaoming Zhu, Sidi Lu, Lei Zheng, Jiaxian Guo, Weinan Zhang, Jun Wang, and Yong Yu. 2018. [Texygen: A benchmarking platform for text generation models](https://md.research.elamur.ai/free/en/papers/zhu-2018-texygen-benchmarking-platform-text-generation/). In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, SIGIR 2018, Ann Arbor, MI, USA, July 08-12, 2018, p…
65. Yuqing Zhu, Xiang Yu, Manmohan Chandraker, and Yu-Xiang Wang. 2020. Private-knn: Practical differential privacy for computer vision. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13-19, 2020, pages 11851–11859. Computer Vision Foundation /… [find](https://scholar.google.com/scholar?q=Yuqing+Zhu%2C+Xiang+Yu%2C+Manmohan+Chandraker%2C+and+Yu%2DXiang+Wang%2E+2020%2E+Private%2Dknn%3A+Practical+differential+privacy+for+computer+vision%2E+In+2020+IEEE%2FCVF+Conference+on+Computer+Vision+and+Pattern+Recognit)
66. Chulin Xie, Zinan Lin, Arturs Backurs, Sivakanth Gopi, Da Yu, Huseyin A. Inan, Harsha Nori, Haotian Jiang, Huishuai Zhang, Yin Tat Lee, Bo Li, and Sergey Yekhanin. 2024. Differentially private synthetic data [find](https://scholar.google.com/scholar?q=Chulin+Xie%2C+Zinan+Lin%2C+Arturs+Backurs%2C+Sivakanth+Gopi%2C+Da+Yu%2C+Huseyin+A%2E+Inan%2C+Harsha+Nori%2C+Haotian+Jiang%2C+Huishuai+Zhang%2C+Yin+Tat+Lee%2C+Bo+Li%2C+and+Sergey+Yekhanin%2E+2024%2E+Differentially+private+synthet)

*Updated 2 November 2026*