Challenge: Recent research shows textual data alone may contain enough information about users' private-attributes that they do not want to disclose such as age, gender, location, political views and sexual orientation.
Approach: They propose a novel Reinforcement Learning-based Text Anonymizor which extracts a latent representation of the original text w.r.t. a given task and leverages deep reinforcement learning to learn an optimal strategy for manipulating text representations w/ the received privacy and utility feedback.
Outcome: The proposed approach preserves both privacy and utility of textual data while preserving its utility.

Similar Papers

Privacy-R1: Privacy-Aware Multi-LLM Agent Collaboration via Reinforcement Learning (2026.acl-long)

Copied to clipboard

Challenge: Prior approaches to rewriting large language models shatters linguistic coherence and removes privacy-sensitive information.
Approach: They propose a framework that trains an agent to dynamically route text chunks . it implicitly distinguishes between replaceable Personally Identifiable Information (PII) and task-critical PII .
Outcome: The proposed framework achieves state-of-the-art on the privacy-utility frontier . it trains an agent to dynamically route text chunks, learning a policy that balances privacy leakage and task performance.
Robust Utility-Preserving Text Anonymization Based on Large Language Models (2025.acl-long)

Copied to clipboard

Challenge: Existing techniques face challenges of re-identification ability of large language models . anonymizing text that contains sensitive information is crucial for a wide range of applications .
Approach: They propose a framework that integrates three key LLM components to perform anonymization.
Outcome: The proposed model outperforms baselines while maintaining greater data utility in downstream tasks.
Anonymisation Models for Text Data: State of the art, Challenges and Future Directions (2021.acl-long)

Copied to clipboard

Challenge: a paper examines the problem of automated text anonymisation . text anonymization is a prerequisite for secure sharing of documents containing sensitive information about individuals.
Approach: They propose to incorporate explicit measures of disclosure risk into the text anonymisation process to reduce the risk of errors.
Outcome: The proposed approach is based on a case study in which the authors outline the benefits and limitations of the proposed methods.
Keep it Private: Unsupervised Privatization of Online Text (2024.naacl-long)

Copied to clipboard

Challenge: Authorship obfuscation has been evaluated in narrow settings in the NLP literature . superficial edit operations can lead to unnatural outputs, authors say .
Approach: They propose an automatic text privatization framework that fine-tunes a large language model via reinforcement learning to produce rewrites that balance soundness, sense, and privacy.
Outcome: The proposed method maintains high text quality according to automated metrics and human evaluation, and successfully evades several automated authorship attacks.
Adaptive Text Anonymization: Learning Privacy-Utility Trade-offs via Prompt Optimization (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for anonymizing textual documents lack flexibility to adapt to diverse requirements.
Approach: They propose a task formulation in which anonymization strategies are automatically adapted to specific privacy–utility requirements.
Outcome: The proposed framework achieves better privacy–utility trade-off than existing baselines on open-source language models while remaining computationally efficient and effective on larger closed-source models.
Privacy-preserving Neural Representations of Text (D18-1)

Copied to clipboard

Challenge: a specific type of attack is used to characterize the privacy of neural representations for NLP tasks, in the context of privacy protection.
Approach: They propose several defense methods based on modified training objectives and characterize the tradeoff between privacy and the utility of neural representations.
Outcome: The proposed defenses improve the privacy of neural representations and characterize the tradeoff between privacy and utility of representations.
Preserving Privacy Through Dememorization: An Unlearning Technique For Mitigating Memorization Risks In Language Models (2023.emnlp-main)

Copied to clipboard

Challenge: Large Language models (LLMs) are trained on vast amounts of data, including sensitive information that poses a risk to personal privacy if exposed.
Approach: They propose a novel unlearning approach that utilizes an efficient reinforcement learning feedback loop via proximal policy optimization to incentivize the LLMs to learn a paraphrasing policy to unlearn the pre-training data.
Outcome: The proposed approach surpasses strong baselines and state-of-the-art methods in terms of its ability to generalize and strike a balance between privacy and LLM performance.
ER-AE: Differentially Private Text Generation for Authorship Anonymization (2021.naacl-main)

Copied to clipboard

Challenge: Recent studies on privacy protection for textual data focus on removing explicit sensitive identifiers without considering the author's writing style.
Approach: They propose a text generation model with an exponential mechanism for authorship anonymization that augments the semantic information through a REINFORCE training reward function.
Outcome: The proposed model outperforms the state-of-the-art on semantic preservation, authorship obfuscation, and stylometric transformation on the real-life peer reviews and Yelp review datasets.
TextFusion: Privacy-Preserving Pre-trained Model Inference via Token Fusion (2022.emnlp-main)

Copied to clipboard

Challenge: Existing methods to preserve inference privacy are available as cloud services . however, the risk of privacy leakage remains, according to recent studies .
Approach: They propose a method to preserve inference privacy by fusing token representations in the cloud.
Outcome: The proposed method preserves inference privacy without sacrificing performance on different scenarios.
CAPE: Context-Aware Private Embeddings for Private Language Learning (2021.emnlp-main)

Copied to clipboard

Challenge: Existing methods to obtain text representations or embeddings with these models encoding personally identifiable information may lead to privacy leaks.
Approach: They propose a novel approach which combines differential privacy and adversarial learning to preserve privacy during training of embeddings.
Outcome: The proposed approach reduces private information leakage by 3% over the current method.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations