Buchumschlag

Beyond Gradient and Priors in Privacy Attacks: Leveraging Pooler Layer Inputs of Language Models in Federated Learning

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Li, Jianwei, Liu, Sheng, Lei, Qi
Format:	Preprint
Veröffentlicht:	2023
Schlagworte:	Machine Learning Artificial Intelligence Computation and Language Cryptography and Security
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

DP-MemArc: Differential Privacy Transfer Learning for Memory Efficient Language Models
von: Liu, Yanming, et al.
Veröffentlicht: (2024)

Privacy-Preserving Data Deduplication for Enhancing Federated Learning of Language Models (Extended Version)
von: Abadi, Aydin, et al.
Veröffentlicht: (2024)

Geminio: Language-Guided Gradient Inversion Attacks in Federated Learning
von: Shan, Junjie, et al.
Veröffentlicht: (2024)

Continuous Embedding Attacks via Clipped Inputs in Jailbreaking Large Language Models
von: Xu, Zihao, et al.
Veröffentlicht: (2024)

Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes
von: Hu, Xiaomeng, et al.
Veröffentlicht: (2024)

Optimal Defenses Against Gradient Reconstruction Attacks
von: Chen, Yuxiao, et al.
Veröffentlicht: (2024)

Learnable Privacy Neurons Localization in Language Models
von: Chen, Ruizhe, et al.
Veröffentlicht: (2024)

Jailbreak Attacks and Defenses Against Large Language Models: A Survey
von: Yi, Sibo, et al.
Veröffentlicht: (2024)

Practical Feasibility of Gradient Inversion Attacks in Federated Learning
von: Valadi, Viktor, et al.
Veröffentlicht: (2025)

Beyond the Tip of Efficiency: Uncovering the Submerged Threats of Jailbreak Attacks in Small Language Models
von: Yi, Sibo, et al.
Veröffentlicht: (2025)

Trustworthy AI: Safety, Bias, and Privacy -- A Survey
von: Fang, Xingli, et al.
Veröffentlicht: (2025)

Privacy Assessment of Federated Learning using Private Personalized Layers
von: Jourdan, Théo, et al.
Veröffentlicht: (2021)

Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning
von: Foroughi, Mohammad Hadi, et al.
Veröffentlicht: (2026)

Jailbreaking LLMs Without Gradients or Priors: Effective and Transferable Attacks
von: Nurlanov, Zhakshylyk, et al.
Veröffentlicht: (2026)

No More Guessing: a Verifiable Gradient Inversion Attack in Federated Learning
von: Diana, Francesco, et al.
Veröffentlicht: (2026)

The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks
von: Chen, Xiaoyi, et al.
Veröffentlicht: (2023)

Fine-Tuning Language Models with Differential Privacy through Adaptive Noise Allocation
von: Li, Xianzhi, et al.
Veröffentlicht: (2024)

Accuracy-Privacy Trade-off in the Mitigation of Membership Inference Attack in Federated Learning
von: Ahamed, Sayyed Farid, et al.
Veröffentlicht: (2024)

Beyond Data Privacy: New Privacy Risks for Large Language Models
von: Du, Yuntao, et al.
Veröffentlicht: (2025)

EIA: Environmental Injection Attack on Generalist Web Agents for Privacy Leakage
von: Liao, Zeyi, et al.
Veröffentlicht: (2024)

The Dark Side of Human Feedback: Poisoning Large Language Models via User Inputs
von: Chen, Bocheng, et al.
Veröffentlicht: (2024)

Forget to Flourish: Leveraging Machine-Unlearning on Pretrained Language Models for Privacy Leakage
von: Rashid, Md Rafi Ur, et al.
Veröffentlicht: (2024)

The Resurgence of GCG Adversarial Attacks on Large Language Models
von: Tan, Yuting, et al.
Veröffentlicht: (2025)

Cutting Through Privacy: A Hyperplane-Based Data Reconstruction Attack in Federated Learning
von: Diana, Francesco, et al.
Veröffentlicht: (2025)

SentinelLMs: Encrypted Input Adaptation and Fine-tuning of Language Models for Private and Secure Inference
von: Mishra, Abhijit, et al.
Veröffentlicht: (2023)

Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation
von: Huang, Tiansheng, et al.
Veröffentlicht: (2025)

PrivacyLens: Evaluating Privacy Norm Awareness of Language Models in Action
von: Shao, Yijia, et al.
Veröffentlicht: (2024)

On Evaluating The Performance of Watermarked Machine-Generated Texts Under Adversarial Attacks
von: Liu, Zesen, et al.
Veröffentlicht: (2024)

Preserving Privacy in Large Language Models: A Survey on Current Threats and Solutions
von: Miranda, Michele, et al.
Veröffentlicht: (2024)

PAL: Proxy-Guided Black-Box Attack on Large Language Models
von: Sitawarin, Chawin, et al.
Veröffentlicht: (2024)

from Benign import Toxic: Jailbreaking the Language Model via Adversarial Metaphors
von: Yan, Yu, et al.
Veröffentlicht: (2025)

FedMentor: Domain-Aware Differential Privacy for Heterogeneous Federated LLMs in Mental Health
von: Sarwar, Nobin, et al.
Veröffentlicht: (2025)

DR-Encoder: Encode Low-rank Gradients with Random Prior for Large Language Models Differentially Privately
von: Wu, Huiwen, et al.
Veröffentlicht: (2024)

Privacy Attack in Federated Learning is Not Easy: An Experimental Study
von: Zhu, Hangyu, et al.
Veröffentlicht: (2024)

MF-CLIP: Leveraging CLIP as Surrogate Models for No-box Adversarial Attacks
von: Zhang, Jiaming, et al.
Veröffentlicht: (2023)

GUIDE: Enhancing Gradient Inversion Attacks in Federated Learning with Denoising Models
von: Carletti, Vincenzo, et al.
Veröffentlicht: (2025)

Local Model Reconstruction Attacks in Federated Learning and their Uses
von: Driouich, Ilias, et al.
Veröffentlicht: (2022)

LLMs can be Dangerous Reasoners: Analyzing-based Jailbreak Attack on Large Language Models
von: Lin, Shi, et al.
Veröffentlicht: (2024)

Context-Aware Membership Inference Attacks against Pre-trained Large Language Models
von: Chang, Hongyan, et al.
Veröffentlicht: (2024)

Federated Foundation Models: Privacy-Preserving and Collaborative Learning for Large Models
von: Yu, Sixing, et al.
Veröffentlicht: (2023)