Search Results for author:
Found 29 papers, 15 papers with code
Mix and Match: Learning-free Controllable Text Generationusing Energy Language Models
Recent work on controlled text generation has either required attribute-based fine-tuning of the base language model (LM), or has restricted the parameterization of the attribute discriminator to be compatible with the base autoregressive LM.
LatticeGen: A Cooperative Framework which Hides Generated Text in a Lattice for Privacy-Aware Generation on Cloud
In the current user-server interaction paradigm of prompted generation with large language models (LLM) on cloud, the server fully controls the generation process, which leaves zero options for users who want to keep the generated text to themselves.
Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation
Our results demonstrate that our algorithm can achieve competitive performance with strong privacy levels.
Membership Inference Attacks against Language Models via Neighbourhood Comparison
To investigate whether this fragility provides a layer of safety, we propose and evaluate neighbourhood attacks, which compare model scores for a given sample to scores of synthetically generated neighbour texts and therefore eliminate the need for access to the training data distribution.
Are Chatbots Ready for Privacy-Sensitive Applications? An Investigation into Input Regurgitation and Prompt-Induced Sanitization
More specifically, we find that when ChatGPT is prompted to summarize cover letters of a 100 candidates, it would retain personally identifiable information (PII) verbatim in 57. 4% of cases, and we find this retention to be non-uniform between different subgroups of people, based on attributes such as gender identity.
Privacy-Preserving Domain Adaptation of Semantic Parsers
Task-oriented dialogue systems often assist users with personal or confidential matters.
Non-Parametric Temporal Adaptation for Social Media Topic Classification
User-generated social media data is constantly changing as new trends influence online discussion and personal information is deleted due to privacy concerns.
Differentially Private Model Compression
Recent papers have shown that large pre-trained language models (LLMs) such as BERT, GPT-2 can be fine-tuned on private data to achieve performance comparable to non-private models for many downstream Natural Language Processing (NLP) tasks while simultaneously guaranteeing differential privacy.
Memorization in NLP Fine-tuning Methods
Large language models are shown to present privacy risks through memorization of training data, and several recent works have studied such risks for the pre-training phase.
FLUTE: A Scalable, Extensible Framework for High-Performance Federated Learning Simulations
We compare the platform with other state-of-the-art platforms and describe available features of FLUTE for experimentation in core areas of active research, such as optimization, privacy, and scalability.
Mix and Match: Learning-free Controllable Text Generation using Energy Language Models
Recent work on controlled text generation has either required attribute-based fine-tuning of the base language model (LM), or has restricted the parameterization of the attribute discriminator to be compatible with the base autoregressive LM.
Quantifying Privacy Risks of Masked Language Models Using Membership Inference Attacks
The wide adoption and application of Masked language models~(MLMs) on sensitive data (from legal to medical) necessitates a thorough quantitative investigation into their privacy vulnerabilities -- to what extent do MLMs leak information about their training data?
What Does it Mean for a Language Model to Preserve Privacy?
Language models lack the ability to understand the context and sensitivity of text, and tend to memorize phrases present in their training sets.
UserIdentifier: Implicit User Representations for Simple and Effective Personalized Sentiment Analysis
As such, these models are often unable to produce personalized responses for individual users, based on their data.
Style Pooling: Automatic Text Style Obfuscation for Improved Classification Fairness
Text style can reveal sensitive attributes of the author (e. g. race or age) to the reader, which can, in turn, lead to privacy violations and bias in both human and algorithmic decisions based on text.
Efficient Hyperparameter Optimization for Differentially Private Deep Learning
Tuning the hyperparameters in the differentially private stochastic gradient descent (DPSGD) is a fundamental challenge.
Benchmarking Differential Privacy and Federated Learning for BERT Models
Natural Language Processing (NLP) techniques can be applied to help with the diagnosis of medical conditions such as depression, using a collection of a person's utterances.
When Differential Privacy Meets Interpretability: A Case Study
Given the increase in the use of personal data for training Deep Neural Networks (DNNs) in tasks such as medical imaging and diagnosis, differentially private training of DNNs is surging in importance and there is a large body of work focusing on providing better privacy-utility trade-off.
DP-SGD vs PATE: Which Has Less Disparate Impact on Model Accuracy?
Recent advances in differentially private deep learning have demonstrated that application of differential privacy, specifically the DP-SGD algorithm, has a disparate impact on different sub-groups in the population, which leads to a significantly high drop-in model utility for sub-populations that are under-represented (minorities), compared to well-represented ones.
Privacy Regularization: Joint Privacy-Utility Optimization in LanguageModels
In this work, we introduce two privacy-preserving regularization methods for training language models that enable joint optimization of utility and privacy through (1) the use of a discriminator and (2) the inclusion of a novel triplet-loss term.
Privacy Regularization: Joint Privacy-Utility Optimization in Language Models
In this work, we introduce two privacy-preserving regularization methods for training language models that enable joint optimization of utility and privacy through (1) the use of a discriminator and (2) the inclusion of a triplet-loss term.
U-Noise: Learnable Noise Masks for Interpretable Image Segmentation
Deep Neural Networks (DNNs) are widely used for decision making in a myriad of critical applications, ranging from medical to societal and even judicial.
WAVEQ: GRADIENT-BASED DEEP QUANTIZATION OF NEURAL NETWORKS THROUGH SINUSOIDAL REGULARIZATION
We show how WaveQ balance compute efficiency and accuracy, and provide a heterogeneous bitwidth assignment for quantization of a large variety of deep networks (AlexNet, CIFAR-10, MobileNet, ResNet-18, ResNet-20, SVHN, and VGG-11) that virtually preserves the accuracy.
Neither Private Nor Fair: Impact of Data Imbalance on Utility and Fairness in Differential Privacy
Deployment of deep learning in different fields and industries is growing day by day due to its performance, which relies on the availability of data and compute.
Privacy in Deep Learning: A Survey
In this survey, we review the privacy concerns brought by deep learning, and the mitigating techniques introduced to tackle these issues.
Not All Features Are Equal: Discovering Essential Features for Preserving Prediction Privacy
We formulate this problem as a gradient-based perturbation maximization method that discovers this subset in the input feature space with respect to the functionality of the prediction model used by the provider.
WaveQ: Gradient-Based Deep Quantization of Neural Networks through Sinusoidal Adaptive Regularization
We show how SINAREQ balance compute efficiency and accuracy, and provide a heterogeneous bitwidth assignment for quantization of a large variety of deep networks (AlexNet, CIFAR-10, MobileNet, ResNet-18, ResNet-20, SVHN, and VGG-11) that virtually preserves the accuracy.
Shredder: Learning Noise Distributions to Protect Inference Privacy
To address this challenge, this paper devises Shredder, an end-to-end framework, that, without altering the topology or the weights of a pre-trained network, learns additive noise distributions that significantly reduce the information content of communicated data while maintaining the inference accuracy.
ReLeQ: A Reinforcement Learning Approach for Deep Quantization of Neural Networks
We show how ReLeQ can balance speed and quality, and provide an asymmetric general solution for quantization of a large variety of deep networks (AlexNet, CIFAR-10, LeNet, MobileNet-V1, ResNet-20, SVHN, and VGG-11) that virtually preserves the accuracy (=< 0. 3% loss) while minimizing the computation and storage cost.
