Shadowgraph
TrendingAbout
Sign in

dzungvpham/awesome-llm4privacy

A curated collection of papers and related projects on using LLMs for privacy.

39

14 commits

updated Oct 8, 2025

See the code

README

Awesome LLMs for Privacy

This is a collection of papers (and projects) related to the use of LLMs for privacy that I curate for my research. The focus is on privacy applications powered by modern, decoder-only LLMs (GPT, Llama, etc.), not on training data privacy (e.g., membership inference attacks, data extraction), and not on pre-GPT LLMs like BERT.

All of these papers are from 2023 onward. I try to include paper links from the official publication venues whenever possible, falling back to OpenReview and then to arXiv. I also try to include the earliest date when a paper was first submitted to arXiv. I categorize the papers based on their goals/domains (e.g., detection vs anonymization) and further add the following tags:

  • CO (constructive) if they are building something new or on top of some other existing work
  • EV (evaluation) if they are evaluating/benchmarking (usually there's a novel accompanying dataset)
  • PO (position) if they are primarily presenting a position or opinion
  • CI if they involve the Contextual Integrity theory.

Please feel free to open an issue or a pull request if you would like to share an interesting paper that uses LLMs for privacy. I will try to read and update the list whenever I have the time.

Table of Contents

Detection

Papers that focus on detecting/assessing privacy leakages/risks or privacy policy violations:

General

Named Entity Recognition

(While NER is not necessarily about sensitive data detection, it's very closely related)

Anonymization

Papers that study the use of LLMs for data anonymization, de-identification, sanitization, authorship obfuscation, etc. (further sub-categorized by the target evaluation domain):

General Evaluation

General Anonymizers

Anonymizers for Chatbot Applications

Anonymizers for Agents

Authorship Obfuscation

Special Use Case: Abstractive Summarization

Miscellaneous

Contributors

dzungvpham

14 commits

Similar repos

michele17284/Awesome-Privacy-Preserving-LLMs

Collection of all the papers talking about/relevant to the topic of privacy-preserving LLMs

41

safr-ai-lab/survey-llm

A survey of privacy problems in Large Language Models (LLMs). Contains summary of the corresponding…

69

ydyjya/Awesome-LLM-Safety

A curated list of safety-related papers, articles, and resources focused on Large Language Models…

1,914

chawins/llm-sp

Papers and resources related to the security and privacy of LLMs 🤖

583

Jeryi-Sun/LLM-and-Law

This repository is dedicated to summarizing papers related to large language models with the field…

333

NY1024/Awesome-Trustworthy-GenAI

8

niconi19/LLM-Conversation-Safety

[NAACL2024] Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey

112

chrisliu298/awesome-llm-unlearning

A resource repository for machine unlearning in large language models

626

dzungvpham/awesome-llm4privacy

A curated collection of papers and related projects on using LLMs for privacy.

39

14 commits

updated Oct 8, 2025

See the code

README

Awesome LLMs for Privacy

This is a collection of papers (and projects) related to the use of LLMs for privacy that I curate for my research. The focus is on privacy applications powered by modern, decoder-only LLMs (GPT, Llama, etc.), not on training data privacy (e.g., membership inference attacks, data extraction), and not on pre-GPT LLMs like BERT.

All of these papers are from 2023 onward. I try to include paper links from the official publication venues whenever possible, falling back to OpenReview and then to arXiv. I also try to include the earliest date when a paper was first submitted to arXiv. I categorize the papers based on their goals/domains (e.g., detection vs anonymization) and further add the following tags:

  • CO (constructive) if they are building something new or on top of some other existing work
  • EV (evaluation) if they are evaluating/benchmarking (usually there's a novel accompanying dataset)
  • PO (position) if they are primarily presenting a position or opinion
  • CI if they involve the Contextual Integrity theory.

Please feel free to open an issue or a pull request if you would like to share an interesting paper that uses LLMs for privacy. I will try to read and update the list whenever I have the time.

Table of Contents

Detection

Papers that focus on detecting/assessing privacy leakages/risks or privacy policy violations:

General

Named Entity Recognition

(While NER is not necessarily about sensitive data detection, it's very closely related)

Anonymization

Papers that study the use of LLMs for data anonymization, de-identification, sanitization, authorship obfuscation, etc. (further sub-categorized by the target evaluation domain):

General Evaluation

General Anonymizers

Anonymizers for Chatbot Applications

Anonymizers for Agents

Authorship Obfuscation

Special Use Case: Abstractive Summarization

Miscellaneous

Similar repos

michele17284/Awesome-Privacy-Preserving-LLMs

Collection of all the papers talking about/relevant to the topic of privacy-preserving LLMs

41

safr-ai-lab/survey-llm

A survey of privacy problems in Large Language Models (LLMs). Contains summary of the corresponding…

69

ydyjya/Awesome-LLM-Safety

A curated list of safety-related papers, articles, and resources focused on Large Language Models…

1,914

chawins/llm-sp

Papers and resources related to the security and privacy of LLMs 🤖

583

Jeryi-Sun/LLM-and-Law

This repository is dedicated to summarizing papers related to large language models with the field…

333

NY1024/Awesome-Trustworthy-GenAI

8

niconi19/LLM-Conversation-Safety

[NAACL2024] Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey

112

chrisliu298/awesome-llm-unlearning

A resource repository for machine unlearning in large language models

626

Contributors

dzungvpham

14 commits