Privacy dilemmas and opportunities in large language models: a brief review

Abstract

The growing number of cases indicates that large language model (LLM) brings transformative advancements while raising privacy concerns. Despite promising recent surveys proposed in the literature, there is still a lack of comprehensive analysis dedicated to text privacy specifically for LLM. By comprehensively collecting LLM privacy research, we summarize five privacy issues and their corresponding solutions during both model training and invocation and extend our analysis to three research focuses in LLM application. Moreover, we propose five further research directions and provide prospects for LLM native security mechanisms. Notably, we find that most LLM privacy research is still in the technical exploration phase, with the hope that this work can assist in LLM privacy development.

Graphical abstract

Keywords

large language model / data privacy / data protection

References

[1] Mitchell R. Samsung fab data leak: how ChatGPT exposed sensitive information. See electropages. com/blog/2023/04/how-chatgpt-exposed-sensitive-information website, 2023

[2] G upta M, A kiri C, A ryal K, P arker E, P raharaj L . From ChatGPT to THreatGPT: impact of generative AI in cybersecurity and privacy. IEEE Access, 2023, 11: 80218–80245

[3] Nasr M, Carlini N, Hayase J, Jagielski M, Cooper A F, Ippolito D, Choquette-Choo C A, Wallace E, Tramèr F, Lee K. Scalable extraction of training data from (production) language models. 2023, arXiv preprint arXiv: 2311.17035

[4] Coles C. 11% of data employees paste into ChatGPT is confidential. Cyberhaven. See cyberhaven. com/blog/4–2-of-workers-have-pasted-company-data-into-chatgpt website, 2023

[5] Staab R, Vero M, Balunovic M, Vechev M. Beyond memorization: violating privacy via inference with large language models. In: Proceedings of the 12th International Conference on Learning Representations. 2024

[6] Mehnaz S, Dibbo S V, Kabir E, Li N, Bertino E. Are your sensitive attributes private? Novel model inversion attribute inference attacks on classification models. In: Proceedings of the 31st USENIX Security Symposium (USENIX Security 22). 2022, 4579−4596

[7] Carlini N, Tramèr F, Wallace E, Jagielski M, Herbert-Voss A, Lee K, Roberts A, Brown T B, Song D, Erlingsson Ú, Oprea A, Raffel C. Extracting training data from large language models. In: Bailey M D, Greenstadt R, eds. 30th USENIX Security Symposium. Berkeley: USENIX Association, 2021, 2633−2650

[8] Huang L, Yu W, Ma W, Zhong W, Feng Z, Wang H, Chen Q, Peng W, Feng X, Qin B, Liu T. A survey on hallucination in large language models: principles, taxonomy, challenges, and open questions. 2023, arXiv preprint arXiv: 2311.05232

[9] Hartmann V, Suri A, Bindschaedler V, Evans D, Tople S, West R. SoK: memorization in general-purpose large language models. 2023, arXiv preprint arXiv: 2310.18362

[10] W ang Y, P an Y, Y an M, S u Z, L uan T H . A survey on ChatGPT: AI-generated contents, challenges, and solutions. IEEE Open Journal of the Computer Society, 2023, 4: 280–302

[11] K shetri N . Cybercrime and privacy threats of large language models. IT Professional, 2023, 25( 3): 9–13

[12] Wang T, Zhang Y, Qi S, Zhao R, Xia Z, Weng J. Security and privacy on generative data in AIGC: a survey. ACM Computing Surveys, 2024

[13] Mozes M, He X, Kleinberg B, Griffin L D. Use of LLMs for illicit purposes: threats, prevention measures, and vulnerabilities. 2023, arXiv preprint arXiv: 2308.12833

[14] Smith V, Shamsabadi A S, Ashurst C, Weller A. Identifying and mitigating privacy risks stemming from language models: a survey. 2023, arXiv preprint arXiv: 2310.01424

[15] Y ao Y, D uan J, X u K, C ai Y, S un Z, Z hang Y . A survey on large language model (LLM) security and privacy: the good, the bad, and the ugly. High-Confidence Computing, 2024, 4( 2): 100211

[16] Iqbal U, Kohno T, Roesner F. LLM platform security: applying a systematic evaluation framework to OpenAI’s ChatGPT plugins. In: Proceedings of the 7th AAAI/ACM Conference on AI, Ethics, and Society. 2024, 611−623

[17] Brown H, Lee K, Mireshghallah F, Shokri R, Tramèr F. What does it mean for a language model to preserve privacy? In: Proceedings of 2022 ACM Conference on Fairness, Accountability, and Transparency. 2022, 2280−2292

[18] Z hang D, F inckenberg-Broman P, H oang T, P an S, X ing Z, S taples M, X u X . Right to be forgotten in the era of large language models: implications, challenges, and solutions. AI and Ethics, 2024, 1–10

[19] Neel S, Chang P. Privacy issues in large language models: a survey. 2023, arXiv preprint arXiv: 2312.06717

[20] Das B C, Amini M H, Wu Y. Security and privacy challenges of large language models: a survey. 2024, arXiv preprint arXiv: 2402.00888

[21] Ye J, Chen X, Xu N, Zu C, Shao Z, Liu S, Cui Y, Zhou Z, Gong C, Shen Y, Zhou J, Chen S, Gui T, Zhang Q, Huang X. A comprehensive capability analysis of GPT-3 and GPT-3.5 series models. 2023, arXiv preprint arXiv: 2303.10420

[22] Hoffmann J, Borgeaud S, Mensch A, Buchatskaya E, Cai T, Rutherford E, de Las Casas D, Hendricks L A, Welbl J, Clark A, Hennigan T, Noland E, Millican K, van den Driessche G, Damoc B, Guy A, Osindero S, Simonyan K, Elsen E, Vinyals O, Rae J W, Sifre L. An empirical analysis of compute-optimal large language model training. In: Proceedings of the 36th International Conference on Neural Information Processing Systems. 2022, 30016−30030

[23] Yang H, Liu X, Wang C D. FinGPT: open-source financial large language models. 2023, arXiv preprint arXiv: 2306.06031

[24] Li Y, Li Z, Zhang K, Dan R, Jiang S, Zhang Y. ChatDoctor: a medical chat model fine-tuned on a large language model meta-AI (LLaMA) using medical domain knowledge. 2023, arXiv preprint arXiv: 2303.14070

[25] Hu E J, Shen Y, Wallis P, Allen-Zhu Z, Li Y, Wang S, Wang L, Chen W. LoRA: low-rank adaptation of large language models. In: Proceedings of the 10th International Conference on Learning Representations. 2022

[26] Lester B, Al-Rfou R, Constant N. The power of scale for parameter-efficient prompt tuning. In: Proceedings of 2021 Conference on Empirical Methods in Natural Language Processing. 2021, 3045−3059

[27] Zhao Z, Wallace E, Feng S, Klein D, Singh S. Calibrate before use: improving few-shot performance of language models. In: Proceedings of the 38th International Conference on Machine Learning. 2021, 12697−12706

[28] Ouyang L, Wu J, Jiang X, Almeida D, Wainwright C, Mishkin P, Zhang C, Agarwal S, Slama K, Ray A, Schulman J, Hilton J, Kelton F, Miller L, Simens M, Askell A, Welinder P, Christiano P, Leike J, Lowe L. Training language models to follow instructions with human feedback. In: Proceedings of the 36th International Conference on Neural Information Processing Systems. 2022, 27730−27744

[29] Ullah I, Hassan N, Gill S S, Suleiman B, Ahanger T A, Shah Z, Qadir J, Kanhere S S. Privacy preserving large language models: ChatGPT case study based vision and framework. 2023, arXiv preprint arXiv: 2310.12523

[30] Achiam J, Adler S, Agarwal S, Ahmad L, Akkaya I, , . GPT-4 technical report. 2023, arXiv preprint arXiv: 2303.08774

[31] Jayaraman B, Ghosh E, Chase M, Roy S, Dai W, Evans D. Combing for credentials: active pattern extraction from smart reply. In: Proceedings of 2024 IEEE Symposium on Security and Privacy (SP). 2023, 1443−1461

[32] Mireshghallah F, Goyal K, Uniyal A, Berg-Kirkpatrick T, Shokri R. Quantifying privacy risks of masked language models using membership inference attacks. In: Proceedings of 2022 Conference on Empirical Methods in Natural Language Processing. 2022, 8332−8347

[33] Mouhammad N, Daxenberger J, Schiller B, Habernal I. Crowdsourcing on sensitive data with privacy-preserving text rewriting. In: Proceedings of the 17th Linguistic Annotation Workshop (LAW-XVII). 2023, 73−84

[34] Tang X, Shin R, Inan H A, Manoel A, Mireshghallah F, Lin Z, Gopi S, Kulkarni J, Sim R. Privacy-preserving in-context learning with differentially private few-shot generation. In: Proceedings of the 12th International Conference on Learning Representations. 2024

[35] Tang C, Liu Z, Ma C, Wu Z, Li Y, Liu W, Zhu D, Li Q, Li X, Liu T, Fan L. PolicyGPT: automated analysis of privacy policies with large language models. 2023, arXiv preprint arXiv: 2309.10238

[36] Pan S, Hoang T, Zhang D, Xing Z, Xu X, Lu Q, Staples M. Toward the cure of privacy policy reading phobia: automated generation of privacy nutrition labels from privacy policies. 2023, arXiv preprint arXiv: 2306.10923

[37] Chanenson J, Pickering M, Apthorpe N. Automating governing knowledge commons and contextual integrity (GKC-CI) privacy policy annotations with large language models. 2023, arXiv preprint arXiv: 2311.02192

[38] Lukas N, Salem A, Sim R, Tople S, Wutschitz L, Zanella-Béguelin S. Analyzing leakage of personally identifiable information in language models. In: Proceedings of 2013 IEEE Symposium on Security and Privacy (SP). 2023, 346−363

[39] Chen X, Tang S, Zhu R, Yan S, Jin L, Wang Z, Su L, Zhang Z, Wang X, Tang H. The Janus interface: how fine-tuning in large language models amplifies the privacy risks. 2023, arXiv preprint arXiv: 2310.15469

[40] Huang J, Shao H, Chang K C C. Are large pre-trained language models leaking your personal information? In: Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2022. 2022, 2038−2047

[41] Kandpal N, Wallace E, Raffel C. Deduplicating training data mitigates privacy risks in language models. In: Proceedings of the 39th International Conference on Machine Learning. 2022, 10697−10707

[42] Lee K, Ippolito D, Nystrom A, Zhang C, Eck D, Callison-Burch C, Carlini N. Deduplicating training data makes language models better. In: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022, 8424−8445

[43] Zhang C, Ippolito D, Lee K, Jagielski M, Tramèr F, Carlini N. Counterfactual memorization in neural language models. In: Proceedings of the 37th International Conference on Neural Information Processing Systems. 2023, 39321−39362

[44] Yue X, Inan H A, Li X, Kumar G, McAnallen J, Shajari H, Sun H, Levitan D, Sim R. Synthetic text generation with differential privacy: a simple and practical recipe. In: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023, 1321−1342

[45] Majmudar J, Dupuy C, Peris C, Smaili S, Gupta R, Zemel R. Differentially private decoding in large language models. 2022, arXiv preprint arXiv: 2205.13621

[46] Xiao Y, Jin Y, Bai Y, Wu Y, Yang X, Luo X, Yu W, Zhao X, Liu Y, Gu Q, Chen H, Wang W, Cheng W. PrivacyMind: large language models can be contextual privacy protection learners. 2023, arXiv preprint arXiv: 2310.02469

[47] Behnia R, Ebrahimi M R, Pacheco J, Padmanabhan B. EW-Tune: a framework for privately fine-tuning large language models with differential privacy. In: Proceedings of 2022 IEEE International Conference on Data Mining Workshops (ICDMW). 2022, 560−566

[48] West P, Lu X, Holtzman A, Bhagavatula C, Hwang J D, Choi Y. Reflective decoding: beyond unidirectional generation with off-the-shelf language models. In: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021, 1435−1450

[49] Gokaslan A, Cohen V. Openwebtext corpus. See Skylion007.github.io/OpenWebTextCorpus website, 2019

[50] R affel C, S hazeer N, R oberts A, L ee K, N arang S, M atena M, Z hou Y, L i W, L iu P J . Exploring the limits of transfer learning with a unified text-to-text transformer. The Journal of Machine Learning Research, 2020, 21( 1): 140

[51] Beyer A, Kauermann G, Schütze H. Embedding space correlation as a measure of domain similarity. In: Proceedings of the 12th Language Resources and Evaluation Conference. 2020, 2431−2439

[52] Chelba C, Mikolov T, Schuster M, Ge Q, Brants T, Koehn P, Robinson T. One billion word benchmark for measuring progress in statistical language modeling. 2013, arXiv preprint arXiv: 1312.3005

[53] Zellers R, Holtzman A, Rashkin H, Bisk Y, Farhadi A, Roesner F, Choi Y. Defending against neural fake news. In: Proceedings of the 33rd International Conference on Neural Information Processing Systems. 2019, 812

[54] Tur G, Hakkani-Tür D, Heck L. What is left to be understood in ATIS? In: Proceedings of 2010 IEEE Spoken Language Technology Workshop. 2010, 19−24

[55] Coucke A, Saade A, Ball A, Bluche T, Caulier A, Leroy D, Doumouro C, Gisselbrecht T, Caltagirone F, Lavril T, Primet M, Dureau J. Snips voice platform: an embedded spoken language understanding system for private-by-design voice interfaces. 2018, arXiv preprint arXiv: 1805.10190

[56] Li J, Ott M, Cardie C. Identifying manipulated offerings on review portals. In: Proceedings of 2013 Conference on Empirical Methods in Natural Language Processing. 2013, 1933−1942

[57] Wang A, Singh A, Michael J, Hill F, Levy O, Bowman S R. Glue: a multi-task benchmark and analysis platform for natural language understanding. In: Proceedings of 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP. 2018, 353−355

[58] Socher R, Perelygin A, Wu J, Chuang J, Manning C D, Ng A Y, Potts C. Recursive deep models for semantic compositionality over a sentiment treebank. In: Proceedings of 2013 Conference on Empirical Methods in Natural Language Processing. 2013, 1631−1642

[59] Williams A, Nangia N, Bowman S R. A broad-coverage challenge corpus for sentence understanding through inference. In: Proceedings of 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers). 2017, 1112−1122

[60] Chalkidis I, Androutsopoulos I, Aletras N. Neural legal judgment prediction in English. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 2019, 4317−4323

[61] Klimt B, Yang Y. The Enron corpus: a new dataset for email classification research. In: Proceedings of the 15th European Conference on Machine Learning. 2004, 217−226

[62] Zhang X, Zhao J, LeCun Y. Character-level convolutional networks for text classification. In: Proceedings of the 28th International Conference on Neural Information Processing Systems. 2015, 649−657

[63] Liu J, Cyphers S, Pasupat P, McGraw I, Glass J. A conversational movie search system based on conditional random fields. In: Proceedings of the 13th Annual Conference of the International Speech Communication Association. 2012, 2454−2457

[64] Villalobos P, Sevilla J, Heim L, Besiroglu T, Hobbhahn M, Ho A. Will we run out of data? An analysis of the Limits of scaling datasets in machine learning. 2022, arXiv preprint arXiv: 2211.04325

[65] Chen J, Yang D. Unlearn what you want to forget: efficient unlearning for LLMs. In: Proceedings of 2023 Conference on Empirical Methods in Natural Language Processing. 2023, 12041−12052

[66] Borkar J. What can we learn from data leakage and unlearning for law? 2023, arXiv preprint arXiv: 2307.10476

[67] Hu Z, Zhang Y, Xiao M, Wang W, Feng F, He X. Exact and efficient unlearning for large language model-based recommendation. 2024, arXiv preprint arXiv: 2404.10327

[68] Ziegler C N, McNee S M, Konstan J A, Lausen G. Improving recommendation lists through topic diversification. In: Proceedings of the 14th International Conference on World Wide Web. 2005, 22−32

[69] H arper F M, K onstan J A . The MovieLens datasets: history and context. ACM Transactions on Interactive Intelligent Systems (TiiS), 2015, 5( 4): 19

[70] Jang J, Yoon D, Yang S, Cha S, Lee M, Logeswaran L, Seo M. Knowledge unlearning for mitigating privacy risks in language models. In: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023, 14389−14408

[71] Wang L, Chen T, Yuan W, Zeng X, Wong K F, Yin H. KGA: a general machine unlearning framework based on knowledge gap alignment. In: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023, 13264−13276

[72] Tuggener D, von Däniken P, Peetz T, Cieliebak M. LEDGAR: a large-scale multi-label corpus for text classification of legal provisions in contracts. In: Proceedings of the 12th Language Resources and Evaluation Conference. 2020, 1235−1241

[73] Cettolo M, Niehues J, Stüker S, Bentivogli L, Federico M. Report on the 11th IWSLT evaluation campaign. In: Proceedings of the 11th International Workshop on Spoken Language Translation: Evaluation Campaign. 2014, 2−17

[74] Zhang S, Dinan E, Urbanek J, Szlam A, Kiela D, Weston J. Personalizing dialogue agents: I have a dog, do you have pets too? In: Proceedings of the 6th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2018, 2204−2213

[75] Maas A, Daly R E, Pham P T, Huang D, Ng A Y, Potts C. Learning word vectors for sentiment analysis. In: Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. 2011, 142−150

[76] Gliwa B, Mochol I, Biesek M, Wawer A. SAMSum corpus: a human-annotated dialogue dataset for abstractive summarization. In: Proceedings of the 2nd Workshop on New Frontiers in Summarization. 2019, 70−79

[77] Zaman K, Choshen L, Srivastava S. Fuse to forget: bias reduction and selective memorization through model fusion. In: Proceedings of 2024 Conference on Empirical Methods in Natural Language Processing. 2024, 18763−18783

[78] Rangel F, Rosso P, Verhoeven B, Daelemans W, Potthast M, Stein B. Overview of the 4th author profiling task at pan 2016: cross-genre evaluations. In: Proceedings of the Working Notes Papers of the CLEF 2016 Evaluation Labs. 2016, 750−784

[79] Pawelczyk M, Neel S, Lakkaraju H. In-context unlearning: language models as few-shot unlearners. In: Proceedings of the 41st International Conference on Machine Learning. 2024

[80] Bourtoule L, Chandrasekaran V, Choquette-Choo C A, Jia H, Travers A, Zhang B, Lie D, Papernot N. Machine unlearning. In: Proceedings of 2021 IEEE Symposium on Security and Privacy (SP). 2021, 141−159

[81] Koch K, Soll M. No matter how you slice it: machine unlearning with SISA comes at the expense of minority classes. In: Proceedings of 2023 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML). 2023, 622−637

[82] Li X L, Liang P. Prefix-tuning: optimizing continuous prompts for generation. In: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021, 4582−4597

[83] Wen R, Wang T, Backes M, Zhang Y, Salem A. Last one standing: a comparative analysis of security and privacy of soft prompt tuning, LoRA, and in-context learning. 2023, arXiv preprint arXiv: 2310.11397

[84] Liu R, Wang T, Cao Y, Xiong L. PreCurious: how innocent pre-trained language models turn into privacy traps. 2024, arXiv preprint arXiv: 2403.09562

[85] Qi X, Zeng Y, Xie T, Chen P Y, Jia R, Mittal P, Henderson P. Fine-tuning aligned language models compromises safety, even when users do not intend to! In: Proceedings of the 12th International Conference on Learning Representations. 2024

[86] Li H, Guo D, Fan W, Xu M, Huang J, Meng F, Song Y. Multi-step jailbreaking privacy attacks on ChatGPT. In: Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2023. 2023, 4138−4153

[87] Deng G, Liu Y, Li Y, Wang K, Zhang Y, Li Z, Wang H, Zhang T, Liu Y. MASTERKEY: automated jailbreaking of large language model Chatbots. In: Proc. ISOC NDSS. 2024

[88] Carlini N, Liu C, Erlingsson Ú, Kos J, Song D. The secret sharer: evaluating and testing unintended memorization in neural networks. In: Proceedings of the 28th USENIX Conference on Security Symposium. 2019, 267−284

[89] Ai4Privacy. PII Masking 300k Dataset. See huggingface.co/datasets/ai4privacy/pii-masking-300k website, 2024

[90] Merity S, Xiong C, Bradbury J, Socher R. Pointer sentinel mixture models. In: Proceedings of the 5th International Conference on Learning Representations. 2016

[91] Kornblith S, Norouzi M, Lee H, Hinton G. Similarity of neural network representations revisited. In: Proceedings of the 36th International Conference on Machine Learning. 2019, 3519−3529

[92] Zhang Y, Ippolito D. Prompts should not be seen as secrets: systematically measuring prompt extraction attack success. 2023, arXiv preprint arXiv: 2307.06865

[93] Duan H, Dziedzic A, Papernot N, Boenisch F. Flocks of stochastic parrots: differentially private prompt learning for large language models. In: Proceedings of the 37th International Conference on Neural Information Processing Systems. 2024, 3358

[94] Z hang X, C hen C, X ie Y, C hen X, Z hang J, X iang Y . A survey on privacy inference attacks and defenses in cloud-based deep neural network. Computer Standards & Interfaces, 2023, 83: 103672

[95] Plant R, Gkatzia D, Giuffrida V. CAPE: Context-Aware Private Embeddings for private language learning. In: Proceedings of 2021 Conference on Empirical Methods in Natural Language Processing. 2021, 7970−7978

[96] Li Y, Tan Z, Liu Y. Privacy-preserving prompt tuning for large language model services. 2023, arXiv preprint arXiv: 2305.06212

[97] Feyisetan O, Balle B, Drake T, Diethe T. Privacy- and utility-preserving textual analysis via calibrated multivariate perturbations. In: Proceedings of the 13th International Conference on Web Search and Data Mining. 2020, 178−186

[98] Zhang Z, Zhang X, Xie W, Lu Y. Responsible task automation: empowering large language models as responsible task Automators. 2023, arXiv preprint arXiv: 2306.01242

[99] Ribeiro B, Rolla V, Santos R. INCOGNITUS: a toolbox for automated clinical notes anonymization. In: Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations. 2023, 187−194

[100] Chen Y, Li T, Liu H, Yu Y. Hide and seek (HaS): a lightweight framework for prompt privacy protection. 2023, arXiv preprint arXiv: 2309.03057

[101] Kan Z, Qiao L, Yu H, Peng L, Gao Y, Li D. Protecting user privacy in remote conversational systems: a privacy-preserving framework based on text sanitization. 2023, arXiv preprint arXiv: 2306.08223

[102] Goyal S, Choudhury A R, Raje S, Chakaravarthy V, Sabharwal Y, Verma A. PoWER-BERT: accelerating BERT inference via progressive word-vector elimination. In: Proceedings of the 37th International Conference on Machine Learning. 2020, 346

[103] Modarressi A, Mohebbi H, Pilehvar M T. AdapLeR: speeding up inference by adaptive length reduction. In: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022, 1−15

[104] Zhou X, Lu J, Gui T, Ma R, Fei Z, Wang Y, Ding Y, Cheung Y, Zhang Q, Huang X J. TextFusion: privacy-preserving pre-trained model inference via token fusion. In: Proceedings of 2022 Conference on Empirical Methods in Natural Language Processing. 2022, 8360−8371

[105] Beckerich M, Plein L, Coronado S. RatGPT: Turning online LLMs into proxies for malware attacks. 2023, arXiv preprint arXiv: 2308.09183

[106] Li Y, Wei F, Zhao J, Zhang C, Zhang H. Rain: your language models can align themselves without finetuning. 2023, arXiv preprint arXiv: 2309.07124

[107] Pisano M, Ly P, Sanders A, Yao B, Wang D, Strzalkowski T, Si M. Bergeron: combating adversarial attacks through a conscience-based alignment framework. 2023, arXiv preprint arXiv: 2312.00029

[108] Cao B, Cao Y, Lin L, Chen J. Defending against alignment-breaking attacks via robustly aligned LLM. In: Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024, 10542−10560

[109] Inan H, Upasani K, Chi J, Rungta R, Iyer K, Mao Y, Tontchev M, Hu Q, Fuller B, Testuggine D, Testuggine M. Llama guard: LLM-based input-output safeguard for human-AI conversations. 2023, arXiv preprint arXiv: 2312.06674

[110] Meta. Llm-guard. See llm-guard.com website, 2024

[111] Microsoft. Presidio analyzer. See pypi. org/project/presidio-analyzer/ website, 2024

[112] Xu Z, Liu Y, Deng G, Li Y, Picek S. LLM jailbreak attack versus defense techniques–a comprehensive study. 2024, arXiv preprint arXiv: 2402.13457

[113] Yang X, Chen K, Zhang W, Liu C, Qi Y, Zhang J, Fang H, Yu N. Watermarking text generated by black-box language models. 2023, arXiv preprint arXiv: 2305.08883

[114] Tu S, Sun Y, Bai Y, Yu J, Hou L, Li J. WaterBench: towards holistic evaluation of watermarks for large language models. In: Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024, 1517−1542

[115] Singh K, Zou J. New evaluation metrics capture quality degradation due to LLM watermarking. 2023, arXiv preprint arXiv: 2312.02382

[116] Li Y, Li Q, Cui L, Bi W, Wang Z, Wang L, Yang L, Shi S, Zhang Y. MAGE: machine-generated text detection in the wild. 2023, arXiv preprint arXiv: 2305.13242

[117] Antoun W, Mouilleron V, Sagot B, Seddah D. Towards a robust detection of language model generated text: is ChatGPT that easy to detect? 2023, arXiv preprint arXiv: 2306.05871

[118] Chen Y, Kang H, Zhai V, Li L, Singh R, Ramakrishnan B. GPT-sentinel: distinguishing human and ChatGPT generated content. 2023, arXiv preprint arXiv: 2305.07969

[119] Sarvazyan A M, González J Á, Rosso P, Franco-Salvador M. Supervised machine-generated text detectors: family and scale matters. In: Proceedings of the 14th International Conference of the CLEF Association on Experimental IR Meets Multilinguality, Multimodality, and Interaction. 2023, 121−132

[120] Bhattacharjee A, Kumarage T, Moraffah R, Liu H. ConDA: contrastive domain adaptation for AI-generated text detection. In: Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2023, 598−610

[121] Liu Y, Zhang Z, Zhang W, Yue S, Zhao X, Cheng X, Zhang Y, Hu H. ArguGPT: evaluating, understanding and identifying argumentative essays generated by GPT models. 2023, arXiv preprint arXiv: 2304.07666

[122] B hattacharjee A, L iu H . Fighting fire with fire: can ChatGPT detect AI-generated text?. ACM SIGKDD Explorations Newsletter, 2023, 25( 2): 14–21

[123] Koike R, Kaneko M, Okazaki N. OUTFOX: LLM-generated essay detection through in-context learning with adversarially generated examples. In: Proceedings of the 38th AAAI Conference on Artificial Intelligence. 2024, 21258−21266

[124] Yu X, Qi Y, Chen K, Chen G, Yang X, Zhu P, Zhang W, Yu N. GPT paternity test: GPT generated text detection with GPT genetic inheritance. 2023, arXiv preprint arXiv: 2305.12519

[125] Qiu W, Lie D, Austin L. Calpric: inclusive and fine-grain labeling of privacy policies with crowdsourcing and active learning. In: Proceedings of the 32nd USENIX Conference on Security Symposium. 2023, 60

[126] Cui H, Trimananda R, Markopoulou A, Jordan S. POLIGRAPH: automated privacy policy analysis using knowledge graphs. In: Proceedings of the 32nd USENIX Conference on Security Symposium. 2023, 59

[127] Chen C, Feng X, Li Y, Lyu L, Zhou J, Zheng X, Yin J. Integration of large language models and federated learning. 2023, arXiv preprint arXiv: 2307.08925

[128] Kuang W, Qian B, Li Z, Chen D, Gao D, Pan X, Xie Y, Li Y, Ding B, Zhou J. FederatedScope-LLM: a comprehensive package for fine-tuning large language models in federated learning. In: Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2024, 5260−5271

[129] Gupta S, Huang Y, Zhong Z, Gao T, Li K, Chen D. Recovering private text in federated learning of language models. In: Proceedings of the 36th International Conference on Neural Information Processing Systems. 2022, 8130−8143

[130] Xiao G, Lin J, Han S. Offsite-tuning: transfer learning without full model. 2023, arXiv preprint arXiv: 2302.04870

[131] Fan T, Kang Y, Ma G, Chen W, Wei W, Fan L, Yang Q. Fate-LLM: a industrial grade federated learning framework for large language models. 2023, arXiv preprint arXiv: 2310.10049

[132] Gong Y. Multilevel large language models for everyone. 2023, arXiv preprint arXiv: 2307.13221

[133] Xi Z, Chen W, Guo X, He W, Ding Y, , . The rise and potential of large language model based agents: a survey. 2023, arXiv preprint arXiv: 2309.07864

[134] Wu Q, Bansal G, Zhang J, Wu Y, Zhang S, Zhu E, Li B, Jiang L, Zhang X, Wang C. AutoGen: enabling next-gen LLM applications via multi-agent conversation. 2023, arXiv preprint arXiv: 2308.08155

[135] Zhang H, Du W, Shan J, Zhou Q, Du Y, Tenenbaum J B, Shu T, Gan C. Building cooperative embodied agents modularly with large language models. In: Proceedings of the 12th International Conference on Learning Representations. 2024

[136] Liu X, Lai H, Yu H, Xu Y, Zeng A, Du Z, Zhang P, Dong Y, Tang J. WebGLM: towards an efficient web-enhanced question answering system with human preferences. In: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2023, 4549−4560

[137] Zeng S, Zhang J, He P, Liu Y, Xing Y, Xu H, Ren J, Chang Y, Wang S, Yin D, Tang J. The good and the bad: Exploring privacy issues in retrieval-augmented generation (RAG). In: Proceedings of the Findings of the Association for Computational Linguistics: ACL 2024. 2024, 4505−4524

[138] Chen Y, Mendes E, Das S, Xu W, Ritter A. Can language models be instructed to protect personal information? 2023, arXiv preprint arXiv: 2310.02224

[139] Niu L, Mirza S, Maradni Z, Pöpper C. CodexLeaks: privacy leaks from code generation language models in GitHub copilot. In: Proceedings of the 32nd USENIX Conference on Security Symposium. 2023, 120

[140] He X, Zannettou S, Shen Y, Zhang Y. You only prompt once: on the capabilities of prompt learning on large language models to tackle toxic content. In: Proceedings of 2024 IEEE Symposium on Security and Privacy (SP). 2024, 770−787

[141] Samson L, Barazani N, Ghebreab S, Asano Y M. Privacy-aware visual language models. 2024, arXiv preprint arXiv: 2405.17423

[142] Caldarella S, Mancini M, Ricci E, Aljundi R. The phantom menace: unmasking privacy leakages in vision-language models. 2024, arXiv preprint arXiv: 2408.01228

[143] Bengio Y, Hu E J. Scaling in the service of reasoning & model-based ML. See yoshuabengio. org/2023/03/21/scaling-in-the-service-of-reasoning-model-based-ml/ website, 2023

RIGHTS & PERMISSIONS

Higher Education Press