Auditing Privacy
No mentions found
This entity hasn't been tracked yet, or Iris is still building its knowledge base.
Related Articles from SNS
Detectability in Diversity: Improved Canary Crafting for Privacy Auditing in One Run
arXiv:2605.27292v2 Announce Type: replace Abstract: Privacy auditing aims to empirically assess privacy leakage in machine learning models using membership inference attacks (MIAs), and to derive lower bounds on differential privacy (DP) parameters. Recent one-run auditing methods address the high cost of standard approaches by relying on a single training run with multiple "canary" points whose inclusion or exclusion must be detected by the auditor. In this work, we study the problem of...
Auditing Privacy in Multi-Tenant RAG under Account Collusion
arXiv:2605.19847v2 Announce Type: replace Abstract: Multi-tenant RAG services often treat the account as the privacy boundary: each account receives an $(\varepsilon_{\text{acc}},\delta_{\text{acc}})$-DP retrieval guarantee against the tenant index. We show that this framing understates leakage under same-index account collusion. For Gaussian noise-then-select retrieval, $k$ coordinated same-tenant accounts compose to joint leakage $\Theta(\sqrt{k}\,\varepsilon_{\text{acc}})$, not...
Echelon: Auditable Aggregate-Only Language-Model Adaptation Across Privacy Boundaries
Announce Type: new Abstract: Cross-organization language-model adaptation increasingly faces hard governance constraints: in many deployments, device-level model state-parameters, activations, optimizer state, and per-device updates-cannot be exported outside an administrative boundary. Existing distributed and federated stacks typically assume cross-site model exchange and then retrofit privacy mechanisms, which complicates compliance and makes auditing brittle. We present Echelon, a...
Impact of Graph Structure on Membership-Inference Risk for Graph Neural Networks
arXiv:2601.17130v2 Announce Type: replace Abstract: Graph neural networks (GNNs) are widely used for tasks such as node classification and link prediction, but their use in sensitive settings raises concerns about training-data leakage. Prior work on privacy leakage in GNNs largely borrows assumptions from non-graph domains, overlooking the role of graph structure. We argue for a graph-specific analysis of privacy risk and study how graph structure affects node-level membership inference.
From Agent Traces to Trust: Evidence Tracing and Execution Provenance in LLM Agents
Announce Type: new Abstract: Large language model (LLM)-based agents increasingly solve complex tasks by interacting with external tools, retrieval systems, memory modules, environments, and other agents. These capabilities expand agent autonomy, but also make agent behavior harder to verify, debug, and audit. Final-answer accuracy alone cannot explain how an output was produced, which evidence supported each claim, whether tool calls were justified, how memory influenced later decisions, or...
Apple's next-gen intelligence: Key features to know in Photos, Safari and other apps
At its annual Worldwide Developers Conference this year, WWDC 2026, Apple unveiled its next-generation software suite. The upcoming updates – which span across iOS 27, iPadOS 27, macOS 27, watchOS 27, visionOS 27 and tvOS 27 – introduce a deeply integrated, privacy-focused ecosystem powered by Apple Intelligence. The centerpiece of this rollout is Siri AI, a completely re-engineered assistant capable of understanding on-screen content and personal context.
Cross-modal linkage risk in clinical vision-language models
arXiv:2606.02276v1 Announce Type: new Abstract: Vision-language models (VLMs) trained on paired chest radiographs and radiology reports learn a shared embedding space that can preserve instance-level image-report correspondence. This poses a privacy risk in settings where radiographs and reports are deliberately kept separate after acquisition, such as image-only data sharing or access-controlled reports, because a de-identified image may be re-linked to its original narrative report through...
Investigating Novice Researchers' Perceptions of Research Privacy Within LLM-Assisted Workflows
arXiv:2606.03248v1 Announce Type: new Abstract: Large Language Model (LLMs)-assisted scholarly workflows introduce critical privacy and intellectual property risks. As a uniquely vulnerable cohort driven by publication pressure and a lack of institutional support, novice researchers rely heavily on public LLMs, compelling them to navigate high-stakes privacy-publication trade-offs. To investigate these concerns, we conducted semi-structured interviews with 44 researchers across diverse...
Elon Musk tries again to escape FTC audits of X data handling
Critics hope to keep Elon Musk from escaping a strict data-privacy order imposed by the Federal Trade Commission (FTC) shortly before he took over Twitter. The FTC order placed restrictions on X's data use for 20 years, while requiring regular independent audits and granting the agency authority to request documents as needed to ensure compliance. The FTC’s action came after Twitter voluntarily disclosed that between May 2013 and September 2019, a coding error accidentally allowed phone...
Wrongful Arrest Exposes Failures in One of the Oldest Police Face-Recognition Tools in the US
A Florida man was wrongfully arrested for attempting to illegally lure a child after police relied on a face-recognition match that was inaccurate, according to a lawsuit filed on Wednesday, even though he lived more than 300 miles from the scene and says he had never set foot in the city where the crime took place. Robert Dillon, a 52-year-old commercial crabber from Fort Myers, was arrested after FACES—a face-recognition system operated by Florida’s Pinellas County Sheriff's Office—matched...