惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

D
Docker
F
Fortinet All Blogs
爱范儿
爱范儿
博客园 - Franky
MyScale Blog
MyScale Blog
罗磊的独立博客
博客园 - 三生石上(FineUI控件)
B
Blog
P
Proofpoint News Feed
IT之家
IT之家
宝玉的分享
宝玉的分享
D
DataBreaches.Net
S
SegmentFault 最新的问题
Microsoft Azure Blog
Microsoft Azure Blog
GbyAI
GbyAI
M
MIT News - Artificial intelligence
L
LangChain Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
月光博客
月光博客
雷峰网
雷峰网
Stack Overflow Blog
Stack Overflow Blog
量子位
V
V2EX
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻

cs.CR updates on arXiv.org

Agentic Vulnerability Reasoning on Windows COM Binaries From Beats to Breaches:How Offensive AI Infers Sensitive User Information from Playlists Undetectable Backdoors in Model Parameters: Hiding Sparse Secrets in High Dimensions When Embedding-Based Defenses Fail: Rethinking Safety in LLM-Based Multi-Agent Systems Token-Efficient Change Detection in LLM APIs Selfie-Capture Dynamics as an Auxiliary Signal Against Deepfakes and Injection Attacks for Mobile Identity Verification Trident: Improving Malware Detection with LLMs and Behavioral Features When Alignment Isn't Enough: Response-Path Attacks on LLM Agents RefusalGuard: Geometry-Preserving Fine-Tuning for Safety in LLMs Checkerboard: A Simple, Effective, Efficient and Learning-free Clean Label Backdoor Attack with Low Poisoning Budget Block-wise Codeword Embedding for Reliable Multi-bit Text Watermarking Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors Enhancing Linux Privilege Escalation Attack Capabilities of Local LLM Agents Defusing the Trigger: Plug-and-Play Defense for Backdoored LLMs via Tail-Risk Intrinsic Geometric Smoothing Evaluating Jailbreaking Vulnerabilities in LLMs Deployed as Assistants for Smart Grid Operations: A Benchmark Against NERC Standards Behavioral Canaries: Auditing Private Retrieved Context Usage in RL Fine-Tuning FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models Text Steganography with Dynamic Codebook and Multimodal Large Language Model An AI Agent Execution Environment to Safeguard User Data TwoHamsters: Benchmarking Multi-Concept Compositional Unsafety in Text-to-Image Models Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD Symbolic Guardrails for Domain-Specific Agents: Stronger Safety and Security Guarantees Without Sacrificing Utility Hardening x402: PII-Safe Agentic Payments via Pre-Execution Metadata Filtering QShield: Securing Neural Networks Against Adversarial Attacks using Quantum Circuits Hijacking Text Heritage: Hiding the Human Signature through Homoglyphic Substitution Like a Hammer, It Can Build, It Can Break: Large Language Model Uses, Perceptions, and Adoption in Cybersecurity Operations on Reddit Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation One Word at a Time: Incremental Completion Decomposition Breaks LLM Safety Measuring and Exploiting Contextual Bias in LLM-Assisted Security Code Review
Privacy-Preserving Directly-Follows Graphs: Balancing Ris...
Gamal Elkoumy, Alisa Pankova, Marlon Dumas · 2020-12-02 · via cs.CR updates on arXiv.org

Process mining techniques enable organizations to analyze business process execution traces in order to identify opportunities for improving their operational performance. Oftentimes, such execution traces contain private information. For example, the execution traces of a healthcare process are likely to be privacy-sensitive. In such cases, organizations need to deploy Privacy-Enhancing Technologies (PETs) to strike a balance between the benefits they get from analyzing these data and the requirements imposed onto them by privacy regulations, particularly that of minimizing re-identification risks when data are disclosed to a process analyst. Among many available PETs, differential privacy stands out for its ability to prevent predicate singling out attacks and its composable privacy guarantees. A drawback of differential privacy is the lack of interpretability of the main privacy parameter it relies upon, namely epsilon. This leads to the recurrent question of how much epsilon is enough? This article proposes a method to determine the epsilon value to be used when disclosing the output of a process mining technique in terms of two business-relevant metrics, namely absolute percentage error metrics capturing the loss of accuracy (a.k.a. utility loss) resulting from adding noise to the disclosed data, and guessing advantage, which captures the increase in the probability that an adversary may guess information about an individual as a result of a disclosure. The article specifically studies the problem of protecting the disclosure of the so-called Directly-Follows Graph (DFGs), which is a process mining artifact produced by most process mining tools. The article reports on an empirical evaluation of the utility-risk trade-offs that the proposed approach achieves on a collection of 13 real-life event logs.