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

腾讯CDC
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
博客园 - 叶小钗
人人都是产品经理
人人都是产品经理
博客园 - 聂微东
The Cloudflare Blog
爱范儿
爱范儿
阮一峰的网络日志
阮一峰的网络日志
WordPress大学
WordPress大学
小众软件
小众软件
博客园 - 三生石上(FineUI控件)
Last Week in AI
Last Week in AI
Jina AI
Jina AI
V
V2EX
罗磊的独立博客
V
Visual Studio Blog
A
About on SuperTechFans
IT之家
IT之家
P
Proofpoint News Feed
B
Blog
博客园 - Franky
Blog — PlanetScale
Blog — PlanetScale
Google DeepMind News
Google DeepMind News
Y
Y Combinator Blog

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
Killing four birds with one Gaussian process: the relatio...
Kathrin Grosse, Michael T. Smith, Michael Backes · 2018-06-06 · via cs.CR updates on arXiv.org

In machine learning (ML) security, attacks like evasion, model stealing or membership inference are generally studied in individually. Previous work has also shown a relationship between some attacks and decision function curvature of the targeted model. Consequently, we study an ML model allowing direct control over the decision surface curvature: Gaussian Process classifiers (GPCs). For evasion, we find that changing GPC's curvature to be robust against one attack algorithm boils down to enabling a different norm or attack algorithm to succeed. This is backed up by our formal analysis showing that static security guarantees are opposed to learning. Concerning intellectual property, we show formally that lazy learning does not necessarily leak all information when applied. In practice, often a seemingly secure curvature can be found. For example, we are able to secure GPC against empirical membership inference by proper configuration. In this configuration, however, the GPC's hyper-parameters are leaked, e.g. model reverse engineering succeeds. We conclude that attacks on classification should not be studied in isolation, but in relation to each other.