SoK: Towards Security and Safety of Edge AI
Not on the map. Why?
Abstract (arXiv)
Advanced AI applications have become increasingly available to a broad audience, e.g., as centrally managed large language models (LLMs). Such centralization is both a risk and a performance bottleneck - Edge AI promises to be a solution to these problems. However, its decentralized approach raises additional challenges regarding security and safety. In this paper, we argue that both of these aspects are critical for Edge AI, and even more so, their integration. Concretely, we survey security and safety threats, summarize existing countermeasures, and collect open challenges as a call for more research in this area.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 59 of its 109 references that are in the map. First posted on arXiv 7 Oct 2024.
Builds on
59 of 109 references.
References · 59
- 2023GPT-4 Technical Reportcited by 18,132
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2013Intriguing properties of neural networkscited by 5,114
- 2016Deep Learning with Differential Privacycited by 2,682
- 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
- 2019Fine-Tuning Language Models from Human Preferencescited by 2,124
- 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,330
- 2023Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!cited by 1,106
- 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
- 2019Deep Leakage from Gradientscited by 1,004
38 more
- 2016Practical Black-Box Attacks against Machine Learningcited by 893
- 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
- 2018How To Backdoor Federated Learningcited by 745
- 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 657
- 2019Federated Learning with Differential Privacy: Algorithms and Performance Analysiscited by 528
- 2018Exploiting Unintended Feature Leakage in Collaborative Learningcited by 509
- 2019Local Model Poisoning Attacks to Byzantine-Robust Federated Learningcited by 479
- 2017Differentially Private Federated Learning: A Client Level Perspectivecited by 451
- 2020Label-Only Membership Inference Attackscited by 371
- 2020iDLG: Improved Deep Leakage from Gradientscited by 367
- 2018Analyzing Federated Learning through an Adversarial Lenscited by 346
- 2017Deep Models Under the GAN: Information Leakage from Collaborative Deep Learningcited by 326
- 2021Unsolved Problems in ML Safetycited by 291
- 2020Data Poisoning Attacks Against Federated Learning Systemscited by 266
- 2019Can You Really Backdoor Federated Learning?cited by 250
- 2019Edge Intelligence: Paving the Last Mile of Artificial Intelligence with Edge Computingcited by 225
- 2020Measuring and Reducing Gendered Correlations in Pre-trained Modelscited by 222
- 2023Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Modelscited by 204
- 2023Simple synthetic data reduces sycophancy in large language modelscited by 174
- 2019Stolen Memories: Leveraging Model Memorization for Calibrated White-Box Membership Inferencecited by 168
- 2019Predictive Biases in Natural Language Processing Models: A Conceptual Framework and Overviewcited by 137
- 2020LDP-Fed: Federated Learning with Local Differential Privacycited by 129
- 2013Hacking Smart Machines with Smarter Ones: How to Extract Meaningful Data from Machine Learning Classifierscited by 128
- 2021Sustainable Modular Debiasing of Language Modelscited by 98
- 2020Communication-Efficient Edge AI: Algorithms and Systemscited by 95
- 2020Improving Image Captioning with Better Use of Captionscited by 84
- 2024SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safetycited by 75
- 2022MPAF: Model Poisoning Attacks to Federated Learning based on Fake Clientscited by 65
- 2024Introducing v0.5 of the AI Safety Benchmark from MLCommonscited by 60
- 2019Reducing Gender Bias in Word-Level Language Models with a Gender-Equalizing Loss Functioncited by 57
- 2018Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inferencecited by 45
- 2022SafeText: A Benchmark for Exploring Physical Safety in Language Modelscited by 41
- 2019SHE: A Fast and Accurate Deep Neural Network for Encrypted Datacited by 36
- 2019A General Framework for Implicit and Explicit Debiasing of Distributional Word Vector Spacescited by 36
- 2021Source Inference Attacks in Federated Learningcited by 31
- 2023Stereotypes and Smut: The (Mis)representation of Non-cisgender Identities by Text-to-Image Modelscited by 26
- 2019Shielding Collaborative Learning: Mitigating Poisoning Attacks through Client-Side Detectioncited by 25
- 2023Goal-Oriented Prompt Attack and Safety Evaluation for LLMscited by 23
50 most cited of 59 All 59 (JSON)
Built on by
Citing papers per year: 2026: 1.
1 paper.