Tootfinder

Protect your privacy – Start now with Signal
 
Autocracies always implement broad surveillance methods in order to identify and punish resistance. Surveillance can take many forms including the capture of your social media posts and email, monitoring your connections to web sites, and preventing the use of private communications through encryption back-doors and other means.
 
Take action now to create ways to communicate privately with your family, friends and colleague…

Dual Protection Ring: User Profiling Via Differential Privacy and Service Dissemination Through Private Information Retrieval
Imdad Ullah, Najm Hassan, Tariq Ahamed Ahangar, Zawar Hussain Shah, Mehregan Mahdavi Andrew Levula
https://arxiv.org/abs/2506.13170

Dual Protection Ring: User Profiling Via Differential Privacy and Service Dissemination Through Private Information Retrieval
User profiling is crucial in providing personalised services, as it relies on analysing user behaviour and preferences to deliver targeted services. This approach enhances user experience and promotes heightened engagement. Nevertheless, user profiling also gives rise to noteworthy privacy considerations due to the extensive tracking and monitoring of personal data, potentially leading to surveillance or identity theft. We propose a dual-ring protection mechanism to protect user privacy by exam…

Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers
Tommaso Green, Martin Gubri, Haritz Puerto, Sangdoo Yun, Seong Joon Oh
https://arxiv.org/abs/2506.15674

Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers
We study privacy leakage in the reasoning traces of large reasoning models used as personal agents. Unlike final outputs, reasoning traces are often assumed to be internal and safe. We challenge this assumption by showing that reasoning traces frequently contain sensitive user data, which can be extracted via prompt injections or accidentally leak into outputs. Through probing and agentic evaluations, we demonstrate that test-time compute approaches, particularly increased reasoning steps, ampl…

Fiducial Matching: Differentially Private Inference for Categorical Data
Ogonnaya Michael Romanus, Younes Boulaguiem, Roberto Molinari
https://arxiv.org/abs/2507.11762

Fiducial Matching: Differentially Private Inference for Categorical Data
The task of statistical inference, which includes the building of confidence intervals and tests for parameters and effects of interest to a researcher, is still an open area of investigation in a differentially private (DP) setting. Indeed, in addition to the randomness due to data sampling, DP delivers another source of randomness consisting of the noise added to protect an individual's data from being disclosed to a potential attacker. As a result of this convolution of noises, in many cases…

MExplore: an entity-based visual analytics approach for medical expertise acquisition
Xiao Pang, Yan Huang, Chang Liu, JiYuan Liu, MingYou Liu
https://arxiv.org/abs/2507.12337

MExplore: an entity-based visual analytics approach for medical expertise acquisition
Acquiring medical expertise is a critical component of medical education and professional development. While existing studies focus primarily on constructing medical knowledge bases or developing learning tools based on the structured, private healthcare data, they often lack methods for extracting expertise from unstructured medical texts. These texts constitute a significant portion of medical literature and offer greater flexibility and detail compared to structured data formats. Furthermore…

Analysis: Private-Label Products Are Driving Plant-Based Retail Sales Growth in Four European Countries https://vegconomist.com/market-and-trends/private-label-products-driving-plant-based-retail-sales-growth-four-eu…

Analysis: Private-Label Products Are Driving Plant-Based Retail Sales Growth in Four European Countries - vegconomist - the vegan business magazine
GFI Europe has analysed previously unpublished Circana data, finding that cheaper private label products led to a growth in plant-based retail sales volumes

Versatile and Fast Location-Based Private Information Retrieval with Fully Homomorphic Encryption over the Torus
Joon Soo Yoo, Taeho Kim, Ji Won Yoon
https://arxiv.org/abs/2506.12761

Versatile and Fast Location-Based Private Information Retrieval with Fully Homomorphic Encryption over the Torus
Location-based services often require users to share sensitive locational data, raising privacy concerns due to potential misuse or exploitation by untrusted servers. In response, we present VeLoPIR, a versatile location-based private information retrieval (PIR) system designed to preserve user privacy while enabling efficient and scalable query processing. VeLoPIR introduces three operational modes-interval validation, coordinate validation, and identifier matching-that support a broad range o…

❓ for privacy-minded folks- anyone have a recommendation for a scale that measures more than weight (ex. muscle mass, fat %) but is also privacy-protecting?
And ideally has an “app” or interface that can track the info, see trends over time, etc.
Many choices online… but I have no idea where they are sending data (or to whom they are selling it).
Most private option of course is a scale that keeps all data local, but there is convenience in an app that easily tracks data, s…

A security researcher said
flaws in a carmaker’s online dealership portal
exposed the private information
and vehicle data of its customers,
and could have allowed hackers to remotely break into any of its customers’ vehicles.
Eaton Zveare, who works as a security researcher at software delivery company Harness,
told TechCrunch the flaw he discovered
allowed the creation of an admin account
that granted “unfettered access” to the unnamed carma…

Security flaws in a carmaker's web portal let one hacker remotely unlock cars from anywhere | TechCrunch
Security researcher Eaton Zveare told TechCrunch that the flaws he discovered in the carmaker's centralized dealer portal exposed vast access to customer and vehicle data. With this access, Zveare said he could remotely take over a customer's account and unlock their cars, and more.

EU Digital Regulation and Guatemala: AI, 5G, and Cybersecurity
Victor Lopez Juarez
https://arxiv.org/abs/2508.08315 https://arxiv.org/pdf/2508.08315…

EU Digital Regulation and Guatemala: AI, 5G, and Cybersecurity
The paper examines how EU rules in AI, 5G, and cybersecurity operate as transnational governance and shape policy in Guatemala. It outlines the AI Act's risk approach, the 5G Action Plan and Security Toolbox, and the cybersecurity regime built on ENISA, NIS2, the Cybersecurity Act, and the Cyber Resilience Act. It traces extraterritorial channels such as the Brussels effect, private standards, supply chain clauses, and data transfer controls. Guatemala specific impacts include SME compliance co…

An Enhanced Privacy-preserving Federated Few-shot Learning Framework for Respiratory Disease Diagnosis
Ming Wang, Zhaoyang Duan, Dong Xue, Fangzhou Liu, Zhongheng Zhang
https://arxiv.org/abs/2507.08050 https://arxiv.org/pdf/2507.08050 https://arxiv.org/html/2507.08050
arXiv:2507.08050v1 Announce Type: new
Abstract: The labor-intensive nature of medical data annotation presents a significant challenge for respiratory disease diagnosis, resulting in a scarcity of high-quality labeled datasets in resource-constrained settings. Moreover, patient privacy concerns complicate the direct sharing of local medical data across institutions, and existing centralized data-driven approaches, which rely on amounts of available data, often compromise data privacy. This study proposes a federated few-shot learning framework with privacy-preserving mechanisms to address the issues of limited labeled data and privacy protection in diagnosing respiratory diseases. In particular, a meta-stochastic gradient descent algorithm is proposed to mitigate the overfitting problem that arises from insufficient data when employing traditional gradient descent methods for neural network training. Furthermore, to ensure data privacy against gradient leakage, differential privacy noise from a standard Gaussian distribution is integrated into the gradients during the training of private models with local data, thereby preventing the reconstruction of medical images. Given the impracticality of centralizing respiratory disease data dispersed across various medical institutions, a weighted average algorithm is employed to aggregate local diagnostic models from different clients, enhancing the adaptability of a model across diverse scenarios. Experimental results show that the proposed method yields compelling results with the implementation of differential privacy, while effectively diagnosing respiratory diseases using data from different structures, categories, and distributions.
toXiv_bot_toot

Well this is fun, #Ohio HB96 seems that it would require me to pay to to collect and store your state issued ID if you want to browse Mastodon via this instance.
That's fine, I don't mind spending money which I don't have to collect users' private data including your home address and full legal name. You're fine with giving effectively a complete stranger acro…

Remember the good old days when you could send an letter to the US Gov't at Fort Collins, Colorado. You'd include a self-addressed stamped envelope and a cassette tape and they would return the cassette with a recording of WWV so that you could set your clocks.
You think that is silly? Is it any more silly than that the maga-klan is trying to make US citizens buy weather (data that was acquired via US assets and taxes) from private dealers such as Accuweather?

Wall Street giants like Blackstone, KKR, and BlackRock are pouring hundreds of billions into AI data centers, creating concerns of "oversupply" and a bubble (Maureen Farrell/New York Times)
https://www.nytimes.com/2025/06/02/business/ai-da…

Are A.I. Data Centers a Sure Thing or the Next Real Estate Bubble?
Private equity firms like Blackstone are using their clients’ money to buy and build data centers to fuel the artificial intelligence boom.

The EU wants to decrypt your private data by 2030 https://www.techradar.com/vpn/vpn-privacy-security/the-eu-wants-to-decrypt-your-private-data-by-2030
The EU Commission unveiled the first step in its security strategy to ensure …

Dual Protection Ring: User Profiling Via Differential Privacy and Service Dissemination Through Private Information Retrieval
Imdad Ullah, Najm Hassan, Tariq Ahamed Ahangar, Zawar Hussain Shah, Mehregan Mahdavi Andrew Levula
https://arxiv.org/abs/2506.13170

Dual Protection Ring: User Profiling Via Differential Privacy and Service Dissemination Through Private Information Retrieval
User profiling is crucial in providing personalised services, as it relies on analysing user behaviour and preferences to deliver targeted services. This approach enhances user experience and promotes heightened engagement. Nevertheless, user profiling also gives rise to noteworthy privacy considerations due to the extensive tracking and monitoring of personal data, potentially leading to surveillance or identity theft. We propose a dual-ring protection mechanism to protect user privacy by exam…

Maybe it's time for security forces to understand that information security is a) also important and b) easily compromised. The Strava vulnerability is a known issue - I saw news articles about that a few years ago. But surely by now the security world should know that any app that shares information on an unsecure system is a risk?
And I wonder what a review of how this is playing out in the US would show. Inquiring minds, and all that.

Maybe it's time for security forces to understand that information security is a) also important and b) easily compromised. The Strava vulnerability is a known issue - I saw news articles about that a few years ago. But surely by now the security world should know that any app that shares information on an unsecure system is a risk?
And I wonder what a review of how this is playing out in the US would show. Inquiring minds, and all that.

Optimal Debiased Inference on Privatized Data via Indirect Estimation and Parametric Bootstrap
Zhanyu Wang, Arin Chang, Jordan Awan
https://arxiv.org/abs/2507.10746

Optimal Debiased Inference on Privatized Data via Indirect Estimation and Parametric Bootstrap
We design a debiased parametric bootstrap framework for statistical inference from differentially private data. Existing usage of the parametric bootstrap on privatized data ignored or avoided handling the effect of clamping, a technique employed by the majority of privacy mechanisms. Ignoring the impact of clamping often leads to under-coverage of confidence intervals and miscalibrated type I errors of hypothesis tests. The main reason for the failure of the existing methods is the inconsisten…

"Scale AI left public Google Docs with sensitive details about thousands of its contractors, including their private email addresses and whether they were suspected of
'cheating.'"
https://africa.businessinsider.co…

Scale AI exposed sensitive data about clients like Meta and xAI in public Google Docs, BI finds
As Scale AI seeks to reassure customers that their data is secure following Metas $14

The scale of investment and the involvement of governments means ROI must be found (or rather created) now, by any means necessary! Demand already is being forcefully created to justify these expenditures. Business models, regulations and policies/politics are pivoted in lockstep. Aside from all the conceptual, ethical and environmental issues of LLMs and their required infrastructure, these shifts are already also impacting chip/hardware production pipelines and start spelling the end of pe…

samim (@samim@mastodon.social)
Attached: 1 image The AI infrastructure build-out is so gigantic that in the past 6m, it contributed more to the growth of the U.S. economy than /all of consumer spending/ The 'magnificent 7' spent more than $100 billion on data centers and the like in the past three months *alone* As a % of GDP, spending on AI infra has already exceeded spending on telecom and internet infrastructure from the dot-com boom—and it’s still growing. This spending is so big it’s acting as a sort of privat…

Optimizing Federated Learning for Scalable Power-demand Forecasting in Microgrids
Roopkatha Banerjee, Sampath Koti, Gyanendra Singh, Anirban Chakraborty, Gurunath Gurrala, Bhushan Jagyasi, Yogesh Simmhan
https://arxiv.org/abs/2508.08022

Optimizing Federated Learning for Scalable Power-demand Forecasting in Microgrids
Real-time monitoring of power consumption in cities and micro-grids through the Internet of Things (IoT) can help forecast future demand and optimize grid operations. But moving all consumer-level usage data to the cloud for predictions and analysis at fine time scales can expose activity patterns. Federated Learning~(FL) is a privacy-sensitive collaborative DNN training approach that retains data on edge devices, trains the models on private data locally, and aggregates the local models in the…

My great failure in life is that I never put myself in a position to monetise being this embarrassingly incompetent.
#auspol #australia #government

High-Dimensional Differentially Private Quantile Regression: Distributed Estimation and Statistical Inference
Ziliang Shen, Caixing Wang, Shaoli Wang, Yibo Yan
https://arxiv.org/abs/2508.05212

High-Dimensional Differentially Private Quantile Regression: Distributed Estimation and Statistical Inference
With the development of big data and machine learning, privacy concerns have become increasingly critical, especially when handling heterogeneous datasets containing sensitive personal information. Differential privacy provides a rigorous framework for safeguarding individual privacy while enabling meaningful statistical analysis. In this paper, we propose a differentially private quantile regression method for high-dimensional data in a distributed setting. Quantile regression is a powerful an…

A malicious Jira ticket can cause Cursor to exfiltrate secrets from the repository or local file system. But this is not just a problem with Cursor: GitHub MCP connections can also be exploited to expose private repository data, and a vulnerability in GitLab Duo allowed private information to be exposed through automatically rendered HTML code.

AgentFlayer: When a Jira Ticket Can Steal Your Secrets
TL;DR: A 0click attack through a malicious Jira ticket can cause Cursor to exfiltrate secrets from the repository or local file system.

Reversible Video Steganography Using Quick Response Codes and Modified ElGamal Cryptosystem
Ramadhan J. Mstafa
https://arxiv.org/abs/2508.07289 https://arx…

Reversible Video Steganography Using Quick Response Codes and Modified ElGamal Cryptosystem
The rapid transmission of multimedia information has been achieved mainly by recent advancements in the Internet's speed and information technology. In spite of this, advancements in technology have resulted in breaches of privacy and data security. When it comes to protecting private information in today's Internet era, digital steganography is vital. Many academics are interested in digital video because it has a great capability for concealing important data. There have been a vast number of…

Bistochastically private release of longitudinal data
Nicolas Ruiz
https://arxiv.org/abs/2508.10606 https://arxiv.org/pdf/2508.10606

Bistochastically private release of longitudinal data
Although the bulk of the research in privacy and statistical disclosure control is designed for cross-sectional data, i.e. data where individuals are observed at one single point in time, longitudinal data, i.e. individuals observed over multiple periods, are increasingly collected. Such data enhance undoubtedly the possibility of statistical analysis compared to cross-sectional data, but also come with one additional layer of information, individual trajectories, that must remain practically u…

AquiLLM: a RAG Tool for Capturing Tacit Knowledge in Research Groups
Chandler Campbell, Bernie Boscoe, Tuan Do
https://arxiv.org/abs/2508.05648 https://arx…

AquiLLM: a RAG Tool for Capturing Tacit Knowledge in Research Groups
Research groups face persistent challenges in capturing, storing, and retrieving knowledge that is distributed across team members. Although structured data intended for analysis and publication is often well managed, much of a group's collective knowledge remains informal, fragmented, or undocumented--often passed down orally through meetings, mentoring, and day-to-day collaboration. This includes private resources such as emails, meeting notes, training materials, and ad hoc documentation. To…

This is just one example. "MCP" the protocol for "AI agents" is basically without security measures. It's like running random code on your infrastructure and data.
(Original title: GitHub MCP Exploited: Accessing private repositories via MCP)
https://simonwillison.net…

Domestic frontier AI regulation, an IAEA for AI, an NPT for AI, and a US-led Allied Public-Private Partnership for AI: Four institutions for governing and developing frontier AI
Haydn Belfield
https://arxiv.org/abs/2507.06379

Domestic frontier AI regulation, an IAEA for AI, an NPT for AI, and a US-led Allied Public-Private Partnership for AI: Four institutions for governing and developing frontier AI
Compute governance can underpin international institutions for the governance of frontier AI. To demonstrate this I explore four institutions for governing and developing frontier AI. Next steps for compute-indexed domestic frontier AI regulation could include risk assessments and pre-approvals, data centre usage reports, and release gate regulation. Domestic regimes could be harmonized and monitored through an International AI Agency - an International Atomic Energy Agency (IAEA) for AI. This …

Trump administration is launching a new private health tracking system with Big Tech’s help
The system
-- spearheaded by an administration that has already freely shared highly personal data about Americans in ways that have tested legal bounds
-- could put patients’ desires for more convenience at their doctor’s office on a collision course with their expectations that their medical information be kept private.
“There are enormous ethical and legal concerns,”
…

Trump administration announces new private health tracking system with Big Tech’s help
The Trump administration is pushing for Americans to upload personal health data to new apps run by private tech companies. This initiative aims to make accessing health records and monitoring wellness easier. On Wednesday, President Donald Trump is expected to discuss this at an event with leaders from companies like Google and Amazon. The system will focus on diabetes, weight management, and digital tools for patient check-ins. While officials say patients must opt in and data will be secure,…

Membership and Memorization in LLM Knowledge Distillation
Ziqi Zhang, Ali Shahin Shamsabadi, Hanxiao Lu, Yifeng Cai, Hamed Haddadi
https://arxiv.org/abs/2508.07054 https://

Membership and Memorization in LLM Knowledge Distillation
Recent advances in Knowledge Distillation (KD) aim to mitigate the high computational demands of Large Language Models (LLMs) by transferring knowledge from a large ''teacher'' to a smaller ''student'' model. However, students may inherit the teacher's privacy when the teacher is trained on private data. In this work, we systematically characterize and investigate membership and memorization privacy risks inherent in six LLM KD techniques. Using instruction-tuning settings that span seven NLP t…

A Robust Pipeline for Differentially Private Federated Learning on Imbalanced Clinical Data using SMOTETomek and FedProx
Rodrigo Tertulino
https://arxiv.org/abs/2508.10017 https…

A Robust Pipeline for Differentially Private Federated Learning on Imbalanced Clinical Data using SMOTETomek and FedProx
Federated Learning (FL) presents a groundbreaking approach for collaborative health research, allowing model training on decentralized data while safeguarding patient privacy. FL offers formal security guarantees when combined with Differential Privacy (DP). The integration of these technologies, however, introduces a significant trade-off between privacy and clinical utility, a challenge further complicated by the severe class imbalance often present in medical datasets. The research presented…

"DOGE secured the power to view records that contain competitors’ trade secrets, nonpublic details about government contracts, and sensitive regulatory actions or other information."
'Vulnerable': New alarm as Musk's 'God tier access' to damaging data revealed - Raw Story
https://www.rawstory.com/musk-private-data/

Differentially Private Distribution Release of Gaussian Mixture Models via KL-Divergence Minimization
Hang Liu, Anna Scaglione, Sean Peisert
https://arxiv.org/abs/2506.03467

Differentially Private Distribution Release of Gaussian Mixture Models via KL-Divergence Minimization
Gaussian Mixture Models (GMMs) are widely used statistical models for representing multi-modal data distributions, with numerous applications in data mining, pattern recognition, data simulation, and machine learning. However, recent research has shown that releasing GMM parameters poses significant privacy risks, potentially exposing sensitive information about the underlying data. In this paper, we address the challenge of releasing GMM parameters while ensuring differential privacy (DP) guar…

A look at the rise of $5B unicorns, which make up 13% of ~1,600 unicorns but hold $3.5T of $6T total valuation; 17 startups joined the $5B club in H1 2025 (Gené Teare/Crunchbase News)
https://news.crunchbase.com/venture/ultra-unicorn-startups…

The Rise Of The Ultra-Unicorns: $5B+ Startups Are Leading The Private-Market Herd In 2025
Of the nearly 1,600 companies on The Crunchbase Unicorn Board, a small but ultra-elite club of private companies valued at $5 billion or more is growing across multiple metrics, an analysis of the data shows. We take a look at them, their valuations and which sectors — and countries — rank the highest.

The lethal trifecta for #AI agents: private data, untrusted content, and external communication
https://simonwillison.net/2025/Jun/16/the-lethal-trifecta/

Not really sure about the concerns related to data. I'm of the opinion that government should be creating and opening *more* documents and data to all data users, including all AI companies.
(Obviously I don't mean private and meaningfully confidential data. I mean reports, minutes, procedures, basically any non-confidential document obtainable via FOIA.)

OpenAI and UK sign deal to use AI in public services
The US tech firm behind ChatGPT say it will work with the UK government to 'deliver prosperity for all'.

Data privacy experts are calling on the government to urgently tighten regulation around facial recognition technology, amid growing concerns over its unregulated use by police forces and private companies across the UK.
https://www.computing.co.uk/news/202…

Government urged to tighten regulation around facial recognition technology
Data privacy experts are calling on the government to urgently tighten regulation around facial recognition technology, amid growing concerns over its unregulated use by police ...

Identifying Pre-training Data in LLMs: A Neuron Activation-Based Detection Framework
Hongyi Tang, Zhihao Zhu, Yi Yang
https://arxiv.org/abs/2507.16414 http…

Identifying Pre-training Data in LLMs: A Neuron Activation-Based Detection Framework
The performance of large language models (LLMs) is closely tied to their training data, which can include copyrighted material or private information, raising legal and ethical concerns. Additionally, LLMs face criticism for dataset contamination and internalizing biases. To address these issues, the Pre-Training Data Detection (PDD) task was proposed to identify if specific data was included in an LLM's pre-training corpus. However, existing PDD methods often rely on superficial features like …

Incredible if you think about it...
"The bug, when exploited, allows hackers to steal private digital keys from SharePoint servers without needing any credentials to log in. Once in, the hackers can remotely plant malware, and gain access to the files and data stored within"
Big #Microsoft

New zero-day bug in Microsoft SharePoint under widespread attack | TechCrunch
Security researchers say Microsoft customers should take immediate action to defend against the ongoing cyberattacks, and must assume they have already been compromised.

Chameleon: A MatMul-Free Temporal Convolutional Network Accelerator for End-to-End Few-Shot and Continual Learning from Sequential Data
Douwe den Blanken, Charlotte Frenkel
https://arxiv.org/abs/2505.24852

Chameleon: A MatMul-Free Temporal Convolutional Network Accelerator for End-to-End Few-Shot and Continual Learning from Sequential Data
On-device learning at the edge enables low-latency, private personalization with improved long-term robustness and reduced maintenance costs. Yet, achieving scalable, low-power end-to-end on-chip learning, especially from real-world sequential data with a limited number of examples, is an open challenge. Indeed, accelerators supporting error backpropagation optimize for learning performance at the expense of inference efficiency, while simplified learning algorithms often fail to reach acceptab…

Current State in Privacy-Preserving Text Preprocessing for Domain-Agnostic NLP
Abhirup Sinha, Pritilata Saha, Tithi Saha
https://arxiv.org/abs/2508.03204 https://

Current State in Privacy-Preserving Text Preprocessing for Domain-Agnostic NLP
Privacy is a fundamental human right. Data privacy is protected by different regulations, such as GDPR. However, modern large language models require a huge amount of data to learn linguistic variations, and the data often contains private information. Research has shown that it is possible to extract private information from such language models. Thus, anonymizing such private and sensitive information is of utmost importance. While complete anonymization may not be possible, a number of diffe…

Private Memorization Editing: Turning Memorization into a Defense to Strengthen Data Privacy in Large Language Models
Elena Sofia Ruzzetti, Giancarlo A. Xompero, Davide Venditti, Fabio Massimo Zanzotto
https://arxiv.org/abs/2506.10024

Private Memorization Editing: Turning Memorization into a Defense to Strengthen Data Privacy in Large Language Models
Large Language Models (LLMs) memorize, and thus, among huge amounts of uncontrolled data, may memorize Personally Identifiable Information (PII), which should not be stored and, consequently, not leaked. In this paper, we introduce Private Memorization Editing (PME), an approach for preventing private data leakage that turns an apparent limitation, that is, the LLMs' memorization ability, into a powerful privacy defense strategy. While attacks against LLMs have been performed exploiting previou…

Gradient Inversion Attacks on Parameter-Efficient Fine-Tuning
Hasin Us Sami, Swapneel Sen, Amit K. Roy-Chowdhury, Srikanth V. Krishnamurthy, Basak Guler
https://arxiv.org/abs/2506.04453

Gradient Inversion Attacks on Parameter-Efficient Fine-Tuning
Federated learning (FL) allows multiple data-owners to collaboratively train machine learning models by exchanging local gradients, while keeping their private data on-device. To simultaneously enhance privacy and training efficiency, recently parameter-efficient fine-tuning (PEFT) of large-scale pretrained models has gained substantial attention in FL. While keeping a pretrained (backbone) model frozen, each user fine-tunes only a few lightweight modules to be used in conjunction, to fit speci…

Differentially Private Space-Efficient Algorithms for Counting Distinct Elements in the Turnstile Model
Rachel Cummings, Alessandro Epasto, Jieming Mao, Tamalika Mukherjee, Tingting Ou, Peilin Zhong
https://arxiv.org/abs/2505.23682

Differentially Private Space-Efficient Algorithms for Counting Distinct Elements in the Turnstile Model
The turnstile continual release model of differential privacy captures scenarios where a privacy-preserving real-time analysis is sought for a dataset evolving through additions and deletions. In typical applications of real-time data analysis, both the length of the stream $T$ and the size of the universe $|U|$ from which data come can be extremely large. This motivates the study of private algorithms in the turnstile setting using space sublinear in both $T$ and $|U|$. In …

Indoor Sharing in the Mid-Band: A Performance Study of Neutral-Host, Cellular Macro, and Wi-Fi
Joshua Roy Palathinkal, Muhammad Iqbal Rochman, Vanlin Sathya, Mehmet Yavuz, Monisha Ghosh
https://arxiv.org/abs/2506.04974

Indoor Sharing in the Mid-Band: A Performance Study of Neutral-Host, Cellular Macro, and Wi-Fi
Indoor environments present a significant challenge for wireless connectivity, as immense data demand strains traditional solutions. Public Mobile Network Operators (MNOs), utilizing outdoor macro base stations (BSs), suffer from poor signal penetration. Indoor Wi-Fi networks, on the other hand, may face reliability issues due to spectrum contention. Shared spectrum models, particularly the Citizens Broadband Radio Service (CBRS) utilized by private 4G/5G networks, have emerged as a promising a…

LLMs in Coding and their Impact on the Commercial Software Engineering Landscape
Vladislav Belozerov, Peter J Barclay, Askhan Sami
https://arxiv.org/abs/2506.16653

LLMs in Coding and their Impact on the Commercial Software Engineering Landscape
Large-language-model coding tools are now mainstream in software engineering. But as these same tools move human effort up the development stack, they present fresh dangers: 10% of real prompts leak private data, 42% of generated snippets hide security flaws, and the models can even ``agree'' with wrong ideas, a trait called sycophancy. We argue that firms must tag and review every AI-generated line of code, keep prompts and outputs inside private or on-premises deployments, obey emerging safet…

Privacy Disclosure of Similarity in Speech and Language Processing
Tom B\"ackstr\"om, Mohammad Hassan Vali, My Nguyen, Silas Rech
https://arxiv.org/abs/2508.05250 http…

Privacy Disclosure of Similarity in Speech and Language Processing
Speaker, author, and other biometric identification applications often compare a sample's similarity to a database of templates to determine the identity. Given that data may be noisy and similarity measures can be inaccurate, such a comparison may not reliably identify the true identity as the most similar. Still, even the similarity rank based on an inaccurate similarity measure can disclose private information about the true identity. We propose a methodology for quantifying the privacy disc…

The impact of extracurricular education on socioeconomic mobility in Japan: an application of causal machine learning
Yang Qiang
https://arxiv.org/abs/2506.07421

The impact of extracurricular education on socioeconomic mobility in Japan: an application of causal machine learning
This paper explores the socioeconomic impacts of extracurricular education, specifically private tutoring, on social mobility in Japan. Using data from the 2015 National Survey on Social Stratification and Social Mobility (SSM), we employed a causal machine learning approach to evaluate this educational intervention on income, educational attainment, and occupational prestige. Our research suggests that while shadow education holds the potential for positive socioeconomic impacts, its benefits …

This https://arxiv.org/abs/2406.15119 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSD_…

Breaking Resource Barriers in Speech Emotion Recognition via Data Distillation
Speech emotion recognition (SER) plays a crucial role in human-computer interaction. The emergence of edge devices in the Internet of Things (IoT) presents challenges in constructing intricate deep learning models due to constraints in memory and computational resources. Moreover, emotional speech data often contains private information, raising concerns about privacy leakage during the deployment of SER models. To address these challenges, we propose a data distillation framework to facilitate…

Embedding-Based Federated Data Sharing via Differentially Private Conditional VAEs
Francesco Di Salvo, Hanh Huyen My Nguyen, Christian Ledig
https://arxiv.org/abs/2507.02671

Embedding-Based Federated Data Sharing via Differentially Private Conditional VAEs
Deep Learning (DL) has revolutionized medical imaging, yet its adoption is constrained by data scarcity and privacy regulations, limiting access to diverse datasets. Federated Learning (FL) enables decentralized training but suffers from high communication costs and is often restricted to a single downstream task, reducing flexibility. We propose a data-sharing method via Differentially Private (DP) generative models. By adopting foundation models, we extract compact, informative embeddings, re…

Differential Privacy in Machine Learning: From Symbolic AI to LLMs
Francisco Aguilera-Mart\'inez, Fernando Berzal
https://arxiv.org/abs/2506.11687 http…

Differential Privacy in Machine Learning: From Symbolic AI to LLMs
Machine learning models should not reveal particular information that is not otherwise accessible. Differential privacy provides a formal framework to mitigate privacy risks by ensuring that the inclusion or exclusion of any single data point does not significantly alter the output of an algorithm, thus limiting the exposure of private information. This survey paper explores the foundational definitions of differential privacy, reviews its original formulations and tracing its evolution through…

A deep dive into Apple TV's privacy features shows that Apple's streaming device is more private than the vast majority of alternatives, save for dumb TVs (Scharon Harding/Ars Technica)
https://arstechnica.com/gadgets/2025/0

Breaking down why Apple TVs are privacy advocates’ go-to streaming device
Using the Apple TV app or an Apple account means giving Apple more data, though.

This https://arxiv.org/abs/2505.05031 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csIR_…

LSRP: A Leader-Subordinate Retrieval Framework for Privacy-Preserving Cloud-Device Collaboration
Cloud-device collaboration leverages on-cloud Large Language Models (LLMs) for handling public user queries and on-device Small Language Models (SLMs) for processing private user data, collectively forming a powerful and privacy-preserving solution. However, existing approaches often fail to fully leverage the scalable problem-solving capabilities of on-cloud LLMs while underutilizing the advantage of on-device SLMs in accessing and processing personalized data. This leads to two interconnected…

Necessity of Block Designs for Optimal Locally Private Distribution Estimation
Abigail Gentle
https://arxiv.org/abs/2508.05110 https://arxiv.org/pdf/2508.0…

Necessity of Block Designs for Optimal Locally Private Distribution Estimation
Local differential privacy represents the gold standard for preserving the privacy of data before it leaves the device, and distribution estimation under this model has been well studied. Recently, protocols built upon balanced incomplete block designs were shown to achieve optimal error for this problem. However, it remained unknown whether other constructions could also be optimal. We resolve this question by proving that any protocol achieving optimal error must correspond to some balanced i…

Whatever is left of our constitutional rights and liberties exists mostly in the gaps between what the government knows about us. This is the beginning of the end. Soon, when conservatives are in charge, they'll make us follow their stupid, puritanical rules whether we like it or not, and when liberals are in charge, well, they'll make us follow [i]their[/i] stupid, puritanical rules whether we like it or not.
1/2

U.S. Spy Agencies Are Getting a One-Stop Shop to Buy Your Most Sensitive Personal Data
The government wants to build a centralized platform where spy agencies can more easily buy private info about millions of people.

Aim High, Stay Private: Differentially Private Synthetic Data Enables Public Release of Behavioral Health Information with High Utility
Mohsen Ghasemizade, Juniper Lovato, Christopher M. Danforth, Peter Sheridan Dodds, Laura S. P. Bloomfield, Matthew Price, Team LEMURS, Joseph P. Near
https://arxiv.org/abs/2507.02971

Aim High, Stay Private: Differentially Private Synthetic Data Enables Public Release of Behavioral Health Information with High Utility
Sharing health and behavioral data raises significant privacy concerns, as conventional de-identification methods are susceptible to privacy attacks. Differential Privacy (DP) provides formal guarantees against re-identification risks, but practical implementation necessitates balancing privacy protection and the utility of data. We demonstrate the use of DP to protect individuals in a real behavioral health study, while making the data publicly available and retaining high utility for downst…

Benchmarking and Parallelization of Electrostatic Particle-In-Cell for low-temperature Plasma Simulation by particle-thread Binding
Libn Varghese, Bhaskar Chaudhury, Miral Shah, Mainak Bandyopadhyay
https://arxiv.org/abs/2506.21524

Benchmarking and Parallelization of Electrostatic Particle-In-Cell for low-temperature Plasma Simulation by particle-thread Binding
The Particle-In-Cell (PIC) method for plasma simulation tracks particle phase space information using particle and grid data structures. High computational costs in 2D and 3D device-scale PIC simulations necessitate parallelization, with the Charge Deposition (CD) subroutine often becoming a bottleneck due to frequent particle-grid interactions. Conventional methods mitigate dependencies by generating private grids for each core, but this approach faces scalability issues. We propose a novel ap…

Latent Noise Injection for Private and Statistically Aligned Synthetic Data Generation
Rex Shen, Lu Tian
https://arxiv.org/abs/2506.16636 https://

Latent Noise Injection for Private and Statistically Aligned Synthetic Data Generation
Synthetic Data Generation has become essential for scalable, privacy-preserving statistical analysis. While standard approaches based on generative models, such as Normalizing Flows, have been widely used, they often suffer from slow convergence in high-dimensional settings, frequently converging more slowly than the canonical $1/\sqrt{n}$ rate when approximating the true data distribution. To overcome these limitations, we propose a Latent Noise Injection method using Masked Autoregressive F…

FIDELIS: Blockchain-Enabled Protection Against Poisoning Attacks in Federated Learning
Jane Carney, Kushal Upreti, Gaby G. Dagher, Tim Andersen
https://arxiv.org/abs/2508.10042 …

FIDELIS: Blockchain-Enabled Protection Against Poisoning Attacks in Federated Learning
Federated learning enhances traditional deep learning by enabling the joint training of a model with the use of IoT device's private data. It ensures privacy for clients, but is susceptible to data poisoning attacks during training that degrade model performance and integrity. Current poisoning detection methods in federated learning lack a standardized detection method or take significant liberties with trust. In this paper, we present \Sys, a novel blockchain-enabled poison detection framewor…

Nearly-Linear Time Private Hypothesis Selection with the Optimal Approximation Factor
Maryam Aliakbarpour, Zhan Shi, Ria Stevens, Vincent X. Wang
https://arxiv.org/abs/2506.01162 …

Nearly-Linear Time Private Hypothesis Selection with the Optimal Approximation Factor
Estimating the density of a distribution from its samples is a fundamental problem in statistics. Hypothesis selection addresses the setting where, in addition to a sample set, we are given $n$ candidate distributions -- referred to as hypotheses -- and the goal is to determine which one best describes the underlying data distribution. This problem is known to be solvable very efficiently, requiring roughly $O(\log n)$ samples and running in $\tilde{O}(n)$ time. The quality of the output is mea…

Dependency on Meta AI Chatbot in Messenger Among STEM and Non-STEM Students in Higher Education
Hilene E. Hernandez, Rhiziel P. Manalese, Roque Francis B. Dianelo, Jaymark A. Yambao, Almer B. Gamboa, Lloyd D. Feliciano, Mike Haizon M. David, Freneil R. Pampo, John Paul P. Miranda
https://arxiv.org/abs/2507.21059

Dependency on Meta AI Chatbot in Messenger Among STEM and Non-STEM Students in Higher Education
To understand the potential dependency of tertiary students regarding Meta AI in the academic context. This descriptive cross-sectional study surveyed 872 tertiary students from public and private institutions in Luzon, Philippines. Demographic information and perceptions on Meta AI dependency based on existing literature were collected. Descriptive statistics were used to summarize the data and differences between STEM and non-STEM students were analyzed using the Mann-Whitney U test. The resu…

Israel-based Noma Security, whose platform secures enterprise data and AI models against AI agents, raised a $100M Series B, bringing its total funding to $132M (Steven Scheer/Reuters)
https://www.reuters.com/world/middle-east/

\textit{FedABC}: Attention-Based Client Selection for Federated Learning with Long-Term View
Wenxuan Ye, Xueli An, Junfan Wang, Xueqiang Yan, Georg Carle
https://arxiv.org/abs/2507.20871

\textit{FedABC}: Attention-Based Client Selection for Federated Learning with Long-Term View
Native AI support is a key objective in the evolution of 6G networks, with Federated Learning (FL) emerging as a promising paradigm. FL allows decentralized clients to collaboratively train an AI model without directly sharing their data, preserving privacy. Clients train local models on private data and share model updates, which a central server aggregates to refine the global model and redistribute it for the next iteration. However, client data heterogeneity slows convergence and reduces mo…

Minimax Data Sanitization with Distortion Constraint and Adversarial Inference
Amirarsalan Moatazedian, Yauhen Yakimenka, R\'emi A. Chou, J\"org Kliewer
https://arxiv.org/abs/2507.17942

Minimax Data Sanitization with Distortion Constraint and Adversarial Inference
We study a privacy-preserving data-sharing setting where a privatizer transforms private data into a sanitized version observed by an authorized reconstructor and two unauthorized adversaries, each with access to side information correlated with the private data. The reconstructor is evaluated under a distortion function, while each adversary is evaluated using a separate loss function. The privatizer ensures the reconstructor distortion remains below a fixed threshold while maximizing the mi…

Adapting General-Purpose Embedding Models to Private Datasets Using Keyword-based Retrieval
Yubai Wei, Jiale Han, Yi Yang
https://arxiv.org/abs/2506.00363 …

Adapting General-Purpose Embedding Models to Private Datasets Using Keyword-based Retrieval
Text embedding models play a cornerstone role in AI applications, such as retrieval-augmented generation (RAG). While general-purpose text embedding models demonstrate strong performance on generic retrieval benchmarks, their effectiveness diminishes when applied to private datasets (e.g., company-specific proprietary data), which often contain specialized terminology and lingo. In this work, we introduce BMEmbed, a novel method for adapting general-purpose text embedding models to private data…

This https://arxiv.org/abs/2505.17226 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Secure and Private Federated Learning: Achieving Adversarial Resilience through Robust Aggregation
Federated Learning (FL) enables collaborative machine learning across decentralized data sources without sharing raw data. It offers a promising approach to privacy-preserving AI. However, FL remains vulnerable to adversarial threats from malicious participants, referred to as Byzantine clients, who can send misleading updates to corrupt the global model. Traditional aggregation methods, such as simple averaging, are not robust to such attacks. More resilient approaches, like the Krum algorithm…

Blind Targeting: Personalization under Third-Party Privacy Constraints
Anya Shchetkina
https://arxiv.org/abs/2507.05175 https://arxiv…

Blind Targeting: Personalization under Third-Party Privacy Constraints
Major advertising platforms recently increased privacy protections by limiting advertisers' access to individual-level data. Instead of providing access to granular raw data, the platforms only allow a limited number of aggregate queries to a dataset, which is further protected by adding differentially private noise. This paper studies whether and how advertisers can design effective targeting policies within these restrictive privacy preserving data environments. To achieve this, I develop a p…

SOFT: Selective Data Obfuscation for Protecting LLM Fine-tuning against Membership Inference Attacks
Kaiyuan Zhang, Siyuan Cheng, Hanxi Guo, Yuetian Chen, Zian Su, Shengwei An, Yuntao Du, Charles Fleming, Ashish Kundu, Xiangyu Zhang, Ninghui Li
https://arxiv.org/abs/2506.10424

SOFT: Selective Data Obfuscation for Protecting LLM Fine-tuning against Membership Inference Attacks
Large language models (LLMs) have achieved remarkable success and are widely adopted for diverse applications. However, fine-tuning these models often involves private or sensitive information, raising critical privacy concerns. In this work, we conduct the first comprehensive study evaluating the vulnerability of fine-tuned LLMs to membership inference attacks (MIAs). Our empirical analysis demonstrates that MIAs exploit the loss reduction during fine-tuning, making them highly effective in re…

Statistical Inference for Differentially Private Stochastic Gradient Descent
Xintao Xia, Linjun Zhang, Zhanrui Cai
https://arxiv.org/abs/2507.20560 https://

Statistical Inference for Differentially Private Stochastic Gradient Descent
Privacy preservation in machine learning, particularly through Differentially Private Stochastic Gradient Descent (DP-SGD), is critical for sensitive data analysis. However, existing statistical inference methods for SGD predominantly focus on cyclic subsampling, while DP-SGD requires randomized subsampling. This paper first bridges this gap by establishing the asymptotic properties of SGD under the randomized rule and extending these results to DP-SGD. For the output of DP-SGD, we show that th…

Practical and Accurate Local Edge Differentially Private Graph Algorithms
Pranay Mundra, Charalampos Papamanthou, Julian Shun, Quanquan C. Liu
https://arxiv.org/abs/2506.20828

Practical and Accurate Local Edge Differentially Private Graph Algorithms
The rise of massive networks across diverse domains necessitates sophisticated graph analytics, often involving sensitive data and raising privacy concerns. This paper addresses these challenges using local differential privacy (LDP), which enforces privacy at the individual level, where no third-party entity is trusted, unlike centralized models that assume a trusted curator. We introduce novel LDP algorithms for two fundamental graph statistics: k-core decomposition and triangle counting. Our…

Accelerating Private Heavy Hitter Detection on Continual Observation Streams
Rayne Holland
https://arxiv.org/abs/2507.03361 https://a…

Accelerating Private Heavy Hitter Detection on Continual Observation Streams
Differentially private frequency estimation and heavy hitter detection are core problems in the private analysis of data streams. Two models are typically considered: the one-pass model, which outputs results only at the end of the stream, and the continual observation model, which requires releasing private summaries at every time step. While the one-pass model allows more efficient solutions, continual observation better reflects scenarios where timely and ongoing insights are critical. In …

Differentially Private Explanations for Clusters
Amir Gilad, Tova Milo, Kathy Razmadze, Ron Zadicario
https://arxiv.org/abs/2506.05900 https://

Differentially Private Explanations for Clusters
The dire need to protect sensitive data has led to various flavors of privacy definitions. Among these, Differential privacy (DP) is considered one of the most rigorous and secure notions of privacy, enabling data analysis while preserving the privacy of data contributors. One of the fundamental tasks of data analysis is clustering , which is meant to unravel hidden patterns within complex datasets. However, interpreting clustering results poses significant challenges, and often necessitates an…

This https://arxiv.org/abs/2505.12612 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

EPSpatial: Achieving Efficient and Private Statistical Analytics of Geospatial Data
Geospatial data statistics involve the aggregation and analysis of location data to derive the distribution of clients within geospatial. The need for privacy protection in geospatial data analysis has become paramount due to concerns over the misuse or unauthorized access of client location information. However, existing private geospatial data statistics mainly rely on privacy computing techniques such as cryptographic tools and differential privacy, which leads to significant overhead and in…

Countering Privacy Nihilism
Severin Engelmann, Helen Nissenbaum
https://arxiv.org/abs/2507.18253 https://arxiv.org/pdf/2507.18253

Countering Privacy Nihilism
Of growing concern in privacy scholarship is artificial intelligence (AI), as a powerful producer of inferences. Taken to its limits, AI may be presumed capable of inferring "everything from everything," thereby making untenable any normative scheme, including privacy theory and privacy regulation, which rests on protecting privacy based on categories of data - sensitive versus non-sensitive, private versus public. Discarding data categories as a normative anchoring in privacy and data protecti…

Authenticated Private Set Intersection: A Merkle Tree-Based Approach for Enhancing Data Integrity
Zixian Gong, Zhiyong Zheng, Zhe Hu, Kun Tian, Yi Zhang, Zhedanov Oleksiy, Fengxia Liu
https://arxiv.org/abs/2506.04647

Authenticated Private Set Intersection: A Merkle Tree-Based Approach for Enhancing Data Integrity
Private Set Intersection (PSI) enables secure computation of set intersections while preserving participant privacy, standard PSI existing protocols remain vulnerable to data integrity attacks allowing malicious participants to extract additional intersection information or mislead other parties. In this paper, we propose the definition of data integrity in PSI and construct two authenticated PSI schemes by integrating Merkle Trees with state-of-the-art two-party volePSI and multi-party mPSI pr…

Robust and Differentially Private PCA for non-Gaussian data
Minwoo Kim, Sungkyu Jung
https://arxiv.org/abs/2507.15232 https://arxiv.o…

Robust and Differentially Private PCA for non-Gaussian data
Recent advances have sparked significant interest in the development of privacy-preserving Principal Component Analysis (PCA). However, many existing approaches rely on restrictive assumptions, such as assuming sub-Gaussian data or being vulnerable to data contamination. Additionally, some methods are computationally expensive or depend on unknown model parameters that must be estimated, limiting their accessibility for data analysts seeking privacy-preserving PCA. In this paper, we propose a d…

DP-DocLDM: Differentially Private Document Image Generation using Latent Diffusion Models
Saifullah Saifullah, Stefan Agne, Andreas Dengel, Sheraz Ahmed
https://arxiv.org/abs/2508.04208

DP-DocLDM: Differentially Private Document Image Generation using Latent Diffusion Models
As deep learning-based, data-driven information extraction systems become increasingly integrated into modern document processing workflows, one primary concern is the risk of malicious leakage of sensitive private data from these systems. While some recent works have explored Differential Privacy (DP) to mitigate these privacy risks, DP-based training is known to cause significant performance degradation and impose several limitations on standard training procedures, making its direct applicat…

Benchmarking Fraud Detectors on Private Graph Data
Alexander Goldberg, Giulia Fanti, Nihar Shah, Zhiwei Steven Wu
https://arxiv.org/abs/2507.22347 https://…

Benchmarking Fraud Detectors on Private Graph Data
We introduce the novel problem of benchmarking fraud detectors on private graph-structured data. Currently, many types of fraud are managed in part by automated detection algorithms that operate over graphs. We consider the scenario where a data holder wishes to outsource development of fraud detectors to third parties (e.g., vendors or researchers). The third parties submit their fraud detectors to the data holder, who evaluates these algorithms on a private dataset and then publicly communica…

Detect \& Score: Privacy-Preserving Misbehaviour Detection and Contribution Evaluation in Federated Learning
Marvin Xhemrishi, Alexandre Graell i Amat, Bal\'azs Pej\'o
https://arxiv.org/abs/2506.23583

Detect \& Score: Privacy-Preserving Misbehaviour Detection and Contribution Evaluation in Federated Learning
Federated learning with secure aggregation enables private and collaborative learning from decentralised data without leaking sensitive client information. However, secure aggregation also complicates the detection of malicious client behaviour and the evaluation of individual client contributions to the learning. To address these challenges, QI (Pejo et al.) and FedGT (Xhemrishi et al.) were proposed for contribution evaluation (CE) and misbehaviour detection (MD), respectively. QI, however, l…

VFLAIR-LLM: A Comprehensive Framework and Benchmark for Split Learning of LLMs
Zixuan Gu, Qiufeng Fan, Long Sun, Yang Liu, Xiaojun Ye
https://arxiv.org/abs/2508.03097 https://…

VFLAIR-LLM: A Comprehensive Framework and Benchmark for Split Learning of LLMs
With the advancement of Large Language Models (LLMs), LLM applications have expanded into a growing number of fields. However, users with data privacy concerns face limitations in directly utilizing LLM APIs, while private deployments incur significant computational demands. This creates a substantial challenge in achieving secure LLM adaptation under constrained local resources. To address this issue, collaborative learning methods, such as Split Learning (SL), offer a resource-efficient and p…

Testbed and Software Architecture for Enhancing Security in Industrial Private 5G Networks
Song Son Ha, Florian Foerster, Thomas Robert Doebbert, Tim Kittel, Dominik Merli, Gerd Scholl
https://arxiv.org/abs/2507.20873

Testbed and Software Architecture for Enhancing Security in Industrial Private 5G Networks
In the era of Industry 4.0, the growing need for secure and efficient communication systems has driven the development of fifth-generation (5G) networks characterized by extremely low latency, massive device connectivity and high data transfer speeds. However, the deployment of 5G networks presents significant security challenges, requiring advanced and robust solutions to counter increasingly sophisticated cyber threats. This paper proposes a testbed and software architecture to strengthen the…

Development and analysis of a secured VoIP system for surveillance activities
M. Matsive Ali
https://arxiv.org/abs/2507.21038 https://arxiv.org/pdf/2507.21…

Development and analysis of a secured VoIP system for surveillance activities
Since the 1990s, the telephone has been the primary mode of communication. However, Voice over Internet Protocol (VoIP), which is a highly straightforward and affordable form of data transfer, is now becoming an important part of daily communication. VoIP is the technology that makes it possible to send speech and multimedia data packets across either a public or private IP network. However, a cyberattack known as a man-in-the-middle attack poses a serious concern in transferring data through a…

Differentially Private Synthetic Data Release for Topics API Outputs
Travis Dick, Alessandro Epasto, Adel Javanmard, Josh Karlin, Andres Munoz Medina, Vahab Mirrokni, Sergei Vassilvitskii, Peilin Zhong
https://arxiv.org/abs/2506.23855

Differentially Private Synthetic Data Release for Topics API Outputs
The analysis of the privacy properties of Privacy-Preserving Ads APIs is an area of research that has received strong interest from academics, industry, and regulators. Despite this interest, the empirical study of these methods is hindered by the lack of publicly available data. Reliable empirical analysis of the privacy properties of an API, in fact, requires access to a dataset consisting of realistic API outputs; however, privacy concerns prevent the general release of such data to the publ…

DP-RTFL: Differentially Private Resilient Temporal Federated Learning for Trustworthy AI in Regulated Industries
Abhijit Talluri
https://arxiv.org/abs/2505.23813

DP-RTFL: Differentially Private Resilient Temporal Federated Learning for Trustworthy AI in Regulated Industries
Federated Learning (FL) has emerged as a critical paradigm for enabling privacy-preserving machine learning, particularly in regulated sectors such as finance and healthcare. However, standard FL strategies often encounter significant operational challenges related to fault tolerance, system resilience against concurrent client and server failures, and the provision of robust, verifiable privacy guarantees essential for handling sensitive data. These deficiencies can lead to training disruption…

SoK: Can Synthetic Images Replace Real Data? A Survey of Utility and Privacy of Synthetic Image Generation
Yunsung Chung, Yunbei Zhang, Nassir Marrouche, Jihun Hamm
https://arxiv.org/abs/2506.19360

SoK: Can Synthetic Images Replace Real Data? A Survey of Utility and Privacy of Synthetic Image Generation
Advances in generative models have transformed the field of synthetic image generation for privacy-preserving data synthesis (PPDS). However, the field lacks a comprehensive survey and comparison of synthetic image generation methods across diverse settings. In particular, when we generate synthetic images for the purpose of training a classifier, there is a pipeline of generation-sampling-classification which takes private training as input and outputs the final classifier of interest. In this…

Balancing Privacy and Utility in Correlated Data: A Study of Bayesian Differential Privacy
Martin Lange, Patricia Guerra-Balboa, Javier Parra-Arnau, Thorsten Strufe
https://arxiv.org/abs/2506.21308

Balancing Privacy and Utility in Correlated Data: A Study of Bayesian Differential Privacy
Privacy risks in differentially private (DP) systems increase significantly when data is correlated, as standard DP metrics often underestimate the resulting privacy leakage, leaving sensitive information vulnerable. Given the ubiquity of dependencies in real-world databases, this oversight poses a critical challenge for privacy protections. Bayesian differential privacy (BDP) extends DP to account for these correlations, yet current BDP mechanisms indicate notable utility loss, limiting its ad…

This https://arxiv.org/abs/2412.11369 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

PSGraph: Differentially Private Streaming Graph Synthesis by Considering Temporal Dynamics
Streaming graphs are ubiquitous in daily life, such as evolving social networks and dynamic communication systems. Due to the sensitive information contained in the graph, directly sharing the streaming graphs poses significant privacy risks. Differential privacy, offering strict theoretical guarantees, has emerged as a standard approach for private graph data synthesis. However, existing methods predominantly focus on static graph publishing, neglecting the intrinsic relationship between adjac…

This https://arxiv.org/abs/2504.16449 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

From Past to Present: A Survey of Malicious URL Detection Techniques, Datasets and Code Repositories
Malicious URLs persistently threaten the cybersecurity ecosystem, by either deceiving users into divulging private data or distributing harmful payloads to infiltrate host systems. Gaining timely insights into the current state of this ongoing battle holds significant importance. However, existing reviews exhibit 4 critical gaps: 1) Their reliance on algorithm-centric taxonomies obscures understanding of how detection approaches exploit specific modal information channels; 2) They fail to incor…

How to Securely Shuffle? A survey about Secure Shufflers for privacy-preserving computations
Marc Damie, Florian Hahn, Andreas Peter, Jan Ramon
https://arxiv.org/abs/2507.01487

How to Securely Shuffle? A survey about Secure Shufflers for privacy-preserving computations
Ishai et al. (FOCS'06) introduced secure shuffling as an efficient building block for private data aggregation. Recently, the field of differential privacy has revived interest in secure shufflers by highlighting the privacy amplification they can provide in various computations. Although several works argue for the utility of secure shufflers, they often treat them as black boxes; overlooking the practical vulnerabilities and performance trade-offs of existing implementations. This leaves a ce…

Threshold-Protected Searchable Sharing: Privacy Preserving Aggregated-ANN Search for Collaborative RAG
Ruoyang Rykie Guo
https://arxiv.org/abs/2507.17199 https://

Threshold-Protected Searchable Sharing: Privacy Preserving Aggregated-ANN Search for Collaborative RAG
LLM-powered search services have driven data integration as a significant trend. However, this trend's progress is fundamentally hindered, despite the fact that combining individual knowledge can significantly improve the relevance and quality of responses in specialized queries and make AI more professional at providing services. Two key bottlenecks are private data repositories' locality constraints and the need to maintain compatibility with mainstream search techniques, particularly Hierarc…

This https://arxiv.org/abs/2311.16139 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

GNNBleed: Inference Attacks to Unveil Private Edges in Graphs with Realistic Access to GNN Models
Graph Neural Networks (GNNs) have become indispensable tools for learning from graph structured data, catering to various applications such as social network analysis and fraud detection for financial services. At the heart of these networks are the edges, which are crucial in guiding GNN models' predictions. In many scenarios, these edges represent sensitive information, such as personal associations or financial dealings, which require privacy assurance. However, their contributions to GNN mo…

Communication-Efficient Publication of Sparse Vectors under Differential Privacy
Quentin Hillebrand, Vorapong Suppakitpaisarn, Tetsuo Shibuya
https://arxiv.org/abs/2506.20234

Communication-Efficient Publication of Sparse Vectors under Differential Privacy
In this work, we propose a differentially private algorithm for publishing matrices aggregated from sparse vectors. These matrices include social network adjacency matrices, user-item interaction matrices in recommendation systems, and single nucleotide polymorphisms (SNPs) in DNA data. Traditionally, differential privacy in vector collection relies on randomized response, but this approach incurs high communication costs. Specifically, for a matrix with $N$ users, $n$ columns, and $m$ nonzero …

Machine Learning with Privacy for Protected Attributes
Saeed Mahloujifar, Chuan Guo, G. Edward Suh, Kamalika Chaudhuri
https://arxiv.org/abs/2506.19836 htt…

Machine Learning with Privacy for Protected Attributes
Differential privacy (DP) has become the standard for private data analysis. Certain machine learning applications only require privacy protection for specific protected attributes. Using naive variants of differential privacy in such use cases can result in unnecessary degradation of utility. In this work, we refine the definition of DP to create a more general and flexible framework that we call feature differential privacy (FDP). Our definition is simulation-based and allows for both additio…