Planting and Mitigating Memorized Content in Predictive-Text Language Models

by   C. M. Downey, et al.

Language models are widely deployed to provide automatic text completion services in user products. However, recent research has revealed that language models (especially large ones) bear considerable risk of memorizing private training data, which is then vulnerable to leakage and extraction by adversaries. In this study, we test the efficacy of a range of privacy-preserving techniques to mitigate unintended memorization of sensitive user text, while varying other factors such as model size and adversarial conditions. We test both "heuristic" mitigations (those without formal privacy guarantees) and Differentially Private training, which provides provable levels of privacy at the cost of some model performance. Our experiments show that (with the exception of L2 regularization), heuristic mitigations are largely ineffective in preventing memorization in our test suite, possibly because they make too strong of assumptions about the characteristics that define "sensitive" or "private" text. In contrast, Differential Privacy reliably prevents memorization in our experiments, despite its computational and model-performance costs.


Differentially Private Attention Computation

Large language models (LLMs) have had a profound impact on numerous aspe...

Protecting User Privacy in Remote Conversational Systems: A Privacy-Preserving framework based on text sanitization

Large Language Models (LLMs) are gaining increasing attention due to the...

Selective Differential Privacy for Language Modeling

With the increasing adoption of language models in applications involvin...

Privacy Analysis in Language Models via Training Data Leakage Report

Recent advances in neural network based language models lead to successf...

Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation

We study the problem of in-context learning (ICL) with large language mo...

Mitigating Unintended Memorization in Language Models via Alternating Teaching

Recent research has shown that language models have a tendency to memori...

Mitigating Approximate Memorization in Language Models via Dissimilarity Learned Policy

Large Language models (LLMs) are trained on large amounts of data, which...

Please sign up or login with your details

Forgot password? Click here to reset