Cover Image

Breaking SafetyCore: Exploring the Risks of On-Device AI Deployment

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Guyomard, Victor, Mauvisseau, Mathis, Paindavoine, Marie
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Viewing the process of generating counterfactuals as a source of knowledge: a new approach for explaining classifiers
by: Lemaire, Vincent, et al.
Published: (2023)

Games for AI Control: Models of Safety Evaluations of AI Deployment Protocols
by: Griffin, Charlie, et al.
Published: (2024)

Deploying Large AI Models on Resource-Limited Devices with Split Federated Learning
by: Qiang, Xianke, et al.
Published: (2025)

The Missing Half: Unveiling Training-time Implicit Safety Risks Beyond Deployment
by: Zhang, Zhexin, et al.
Published: (2026)

Performative Risk Control: Calibrating Models for Reliable Deployment under Performativity
by: Li, Victor, et al.
Published: (2025)

Efficiently Deploying LLMs with Controlled Risk
by: Zellinger, Michael J., et al.
Published: (2024)

AutoTailor: Automatic and Efficient Adaptive Model Deployment for Diverse Edge Devices
by: Liu, Mengyang, et al.
Published: (2025)

TinyFormer: Efficient Transformer Design and Deployment on Tiny Devices
by: Yang, Jianlei, et al.
Published: (2023)

Black-Box Behavioral Distillation Breaks Safety Alignment in Medical LLMs
by: Jahan, Sohely, et al.
Published: (2025)

Democratizing AI in Africa: FL for Low-Resource Edge Devices
by: Fabila, Jorge, et al.
Published: (2024)

Empirical Guidelines for Deploying LLMs onto Resource-constrained Edge Devices
by: Qin, Ruiyang, et al.
Published: (2024)

Energy-Efficient Vision Transformer Inference for Edge-AI Deployment
by: Amanzhol, Nursultan, et al.
Published: (2025)

Power Hungry Processing: Watts Driving the Cost of AI Deployment?
by: Luccioni, Alexandra Sasha, et al.
Published: (2023)

MNN-LLM: A Generic Inference Engine for Fast Large Language Model Deployment on Mobile Devices
by: Wang, Zhaode, et al.
Published: (2025)

DTMM: Deploying TinyML Models on Extremely Weak IoT Devices with Pruning
by: Han, Lixiang, et al.
Published: (2024)

Breaking the Safety-Capability Tradeoff: Reinforcement Learning with Verifiable Rewards Maintains Safety Guardrails in LLMs
by: Cho, Dongkyu Derek, et al.
Published: (2025)

The Geometry of Alignment Collapse: When Fine-Tuning Breaks Safety
by: Springer, Max, et al.
Published: (2026)

Efficient Deployment of Vision-Language Models on Mobile Devices: A Case Study on OnePlus 13R
by: Guerrero, Pablo Robin, et al.
Published: (2025)

MobileSafetyBench: Evaluating Safety of Autonomous Agents in Mobile Device Control
by: Lee, Juyong, et al.
Published: (2024)

Graph Neural Networks Automated Design and Deployment on Device-Edge Co-Inference Systems
by: Zhou, Ao, et al.
Published: (2024)

MobileLLM-Flash: Latency-Guided On-Device LLM Design for Industry Scale Deployment
by: Huang, Hanxian, et al.
Published: (2026)

Deployment of AI-Assisted Interventions: Capacity Constraints and Noisy Compliance
by: Chan, Carri W., et al.
Published: (2026)

Patient Safety Risks from AI Scribes: Signals from End-User Feedback
by: Dai, Jessica, et al.
Published: (2025)

On-Device Vision Training, Deployment, and Inference on a Thumb-Sized Microcontroller
by: Ellis, Jeremy
Published: (2026)

Breaking Bad: Interpretability-Based Safety Audits of State-of-the-Art LLMs
by: Agarwal, Krishiv, et al.
Published: (2026)

What Makes and Breaks Safety Fine-tuning? A Mechanistic Study
by: Jain, Samyak, et al.
Published: (2024)

A Cost-Benefit Analysis of On-Premise Large Language Model Deployment: Breaking Even with Commercial LLM Services
by: Pan, Guanzhong, et al.
Published: (2025)

Constrained Edge AI Deployment: Fine-Tuning vs Distillation for LLM Compression
by: Sander, Jacob, et al.
Published: (2025)

A New Frontier of AI: On-Device AI Training and Personalization
by: Moon, Ji Joong, et al.
Published: (2022)

NeuroAI for AI Safety
by: Mineault, Patrick, et al.
Published: (2024)

When Alpha Breaks: Two-Level Uncertainty for Safe Deployment of Cross-Sectional Stock Rankers
by: Sanderink, Ursina
Published: (2026)

An Explainable and Fair AI Tool for PCOS Risk Assessment: Calibration, Subgroup Equity, and Interactive Clinical Deployment
by: Khan, Asma Sadia, et al.
Published: (2025)

Edge AI for Automotive Vulnerable Road User Safety: Deployable Detection via Knowledge Distillation
by: Karjol, Akshay, et al.
Published: (2026)

Statistically Valid Post-Deployment Monitoring Should Be Standard for AI-Based Digital Health
by: Dolin, Pavel, et al.
Published: (2025)

Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment
by: Vega, Jason, et al.
Published: (2024)

SymLight: Exploring Interpretable and Deployable Symbolic Policies for Traffic Signal Control
by: Liao, Xiao-Cheng, et al.
Published: (2025)

Prediction-Powered Risk Monitoring of Deployed Models for Detecting Harmful Distribution Shifts
by: Zhang, Guangyi, et al.
Published: (2026)

Towards Sustainable Personalized On-Device Human Activity Recognition with TinyML and Cloud-Enabled Auto Deployment
by: Saha, Bidyut, et al.
Published: (2024)

Breaking Down Power Barriers in On-Device Streaming ASR: Insights and Solutions
by: Li, Yang, et al.
Published: (2024)

Deployment-complete benchmarking
by: Mansouri, El Mustapha, et al.
Published: (2026)