lins-sufe / LLMs-Privacy-Literature

LLMs Privacy Literature

A curated list of LLMs security research papers.

Paper are divided into three categories: motivation, possible dataset and methods design and sorted by their released dates in descending order.

Quick Links

Motivation: Link
Text Anonymization: Link Style Transfer: Link Prompt attack: Link
Jailbreak attack: Link
De-identification and anonymization of medical reports: Link
Prompt Learning Link
Text Generation Link

Motivation [Back to Top]

Year	Title	Conference	Field	Venue	Paper Link	Code Link
2023	Beyond Memorization: Violating Privacy Via Inference with Large Language Models	unknown	unknown	arXiv	Link	unknown
2023	DECODINGTRUST: A Comprehensive Assessment of Trustworthiness in GPT Models	NeurIPS	unknown	arXiv	Link	Link

Text Anonymization [Back to Top]

Year	Title	Conference	Field	Venue	Paper Link	Code Link
2021	ADePT: Auto-encoder based Differentially Private Text Transformation	ACL2021	DP(sentence level)	arxiv	Link	Link
2022	DP-VAE: Human-Readable Text Anonymization for Online Reviews with Differentially Private Variational Autoencoders	WWW’22	DP(sentence level)	ACM	Link	unknown
2022	A Customized Text Sanitization Mechanism with Differential Privacy	ACL2023	DP(token level)	arxiv	Link	Link
2023	Reducing Privacy Risks in Online Self-Disclosures with Language Models	unknown	Finetune	arxiv	Link	unknown
2023	DP-BART for Privatized Text Rewriting under Local Differential Privacy	ACL2023	DP(sentence level)	arxiv	Link	Link
2023	Locally Differentially Private Document Generation UsingZero Shot Prompting	EMNLP	DP(word level)	arxiv	Link	Link
2023	InferDPT: Privacy-preserving Inference for Black-box Large Language Models	unknown	DP(token level)	arxiv	Link	unknown
2023	Differentially Private Natural Language Models: Recent Advances and Future Directions	unknown	survey	arxiv	Link	unknown
2023	DEFENDING AGAINST AUTHORSHIP IDENTIFICATION ATTACKS	unknown	survey	arxiv	Link	unknown

Style Transfer [Back to Top]

Year	Title	Conference	Field	Venue	Paper Link	Code Link
2018	Delete, Retrieve, Generate:A Simple Approach to Sentiment and Style Transfer	ACL	unknown	ACL	Link	Link
2019	Disentangled Representation Learning for Non-Parallel Text Style Transfer	ACL	Disentangle	ACL	Link	Link

Prompt attack [Back to Top]

Year	Title	Conference	Field	Venue	Paper Link	Code Link
2023	Prompt Injection Attacks and Defenses in LLM-Integrated Applications	unknown	unknown	arXiv	Link	Link
2023	PromptBench: Towards Evaluating the Robustness of Large Language Models on Adversarial Prompts	unknown	unknown	arXiv	Link	Link

De-identification and anonymization of medical reports [Back to Top]

Year	Title	Conference	Field	Venue	Paper Link	Code Link
2023	Are Chatbots Ready for Privacy-Sensitive Applications? An Investigation into Input Regurgitation and Prompt-Induced Sanitization	unknown	unknown	arXiv	Link	unknown
2023	DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4	unknown	unknown	arXiv	Link	unknown

Jailbreak attack [Back to Top]

Year	Title	Conference	Field	Venue	Paper Link	Code Link
2023	ON THE SAFETY OF OPEN-SOURCED LARGE LANGUAGE MODELS: DOES ALIGNMENT REALLY PREVENT THEM FROM BEING MISUSED?	unknown	Jailbreak	arXiv	Link	unknown
2023	Do Anything Now”: Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models	unknown	unknown	arXiv	unknown
2023	MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots	unknown	unknown	arXiv	Link	unknown
2023	Universal and Transferable Adversarial Attacks on Aligned Language Models	unknown	unknown	arXiv	Link	unknown
2023	SneakyPrompt: Jailbreaking Text-to-image Generative Models	IEEE S&P 2024	unknown	arXiv	Link	Link
2024	How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs	unknown	unknown	unknown	Link	unknown

Prompt Learning [Back to Top]

Year	Title	Conference	Field	Venue	Paper Link	Code Link
2021	The Power of Scale for Parameter-Efficient Prompt Tuning	unknown	unknown	arXiv	Link	unknown
2023	You Only Prompt Once: On the Capabilities of Prompt Learning on Large Language Models to Tackle Toxic Content?	S&P	unknown	arXiv	Link	Link

Text Generation [Back to Top]

Year	Title	Conference	Type	Venue	Paper Link	Code Link
2021	Pretrained Language Models for Text Generation: A Survey	ACM	survey	arxiv	Link	unknown
2021	Automatic text summarization: A comprehensive survey	Expert Syst. Appl. (2021)	survey	ScienceDirect	Link	unknown
2022	ParaDetox: Detoxification with Parallel Data	ACL	paper	acl	Link	unknown
2023	A Systematic survey on automated text generation tools and techniques: application, evaluation, and challenges	MULTIMED TOOLS APPL	survey	springer	Link	unknown