Imagen de Portada

PureGen: Universal Data Purification for Train-Time Poison Defense via Generative Model Dynamics

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Bhat, Sunay, Jiang, Jeffrey, Pooladzandi, Omead, Branch, Alexander, Pottie, Gregory
Formato:	Preprint
Publicado:	2024
Materias:	Machine Learning Artificial Intelligence Cryptography and Security
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

PureEBM: Universal Poison Purification via Mid-Run Dynamics of Energy-Based Models
por: Pooladzandi, Omead, et al.
Publicado: (2024)

PUREVQ-GAN: Defending Data Poisoning Attacks through Vector-Quantized Bottlenecks
por: Branch, Alexander, et al.
Publicado: (2025)

Defense against Poisoning Attacks under Shuffle-DP
por: Wang, Siyi, et al.
Publicado: (2026)

Dynamic Dual-level Defense Routing for Continual Adversarial Training
por: Wang, Wenxuan, et al.
Publicado: (2025)

A Learning-Based Attack Framework to Break SOTA Poisoning Defenses in Federated Learning
por: Yang, Yuxin, et al.
Publicado: (2024)

SCOUT: A Defense Against Data Poisoning Attacks in Fine-Tuned Language Models
por: Afane, Mohamed, et al.
Publicado: (2025)

Mitigating the Privacy Issues in Retrieval-Augmented Generation (RAG) via Pure Synthetic Data
por: Zeng, Shenglai, et al.
Publicado: (2024)

Poisoning Attacks and Defenses in Recommender Systems: A Survey
por: Wang, Zongwei, et al.
Publicado: (2024)

Data Poisoning for In-context Learning
por: He, Pengfei, et al.
Publicado: (2024)

SecureLearn -- An Attack-agnostic Defense for Multiclass Machine Learning Against Data Poisoning Attacks
por: Paracha, Anum, et al.
Publicado: (2025)

A Data-Driven Defense against Edge-case Model Poisoning Attacks on Federated Learning
por: Purohit, Kiran, et al.
Publicado: (2023)

Mellivora Capensis: A Backdoor-Free Training Framework on the Poisoned Dataset without Auxiliary Data
por: Pu, Yuwen, et al.
Publicado: (2024)

Revisiting the Auxiliary Data in Backdoor Purification
por: Wei, Shaokui, et al.
Publicado: (2025)

Turning Generative Models Degenerate: The Power of Data Poisoning Attacks
por: Jiang, Shuli, et al.
Publicado: (2024)

Memory Poisoning Attack and Defense on Memory Based LLM-Agents
por: Sunil, Balachandra Devarangadi, et al.
Publicado: (2026)

Sharpness-Aware Data Poisoning Attack
por: He, Pengfei, et al.
Publicado: (2023)

Deep-Research Agents Can Be Poisoned via User-Generated Content
por: Zhang, Tingwei, et al.
Publicado: (2026)

Fight Poison with Poison: Enhancing Robustness in Few-shot Machine-Generated Text Detection with Adversarial Training
por: Duan, Wenjing, et al.
Publicado: (2026)

FlowPure: Continuous Normalizing Flows for Adversarial Purification
por: Collaert, Elias, et al.
Publicado: (2025)

Defending against Data Poisoning Attacks in Federated Learning via User Elimination
por: Galanis, Nick
Publicado: (2024)

Defending Against Neural Network Model Inversion Attacks via Data Poisoning
por: Zhou, Shuai, et al.
Publicado: (2024)

Through the Stealth Lens: Attention-Aware Defenses Against Poisoning in RAG
por: Choudhary, Sarthak, et al.
Publicado: (2025)

Dual Defense: Enhancing Privacy and Mitigating Poisoning Attacks in Federated Learning
por: Xu, Runhua, et al.
Publicado: (2025)

LocalAlign: Enabling Generalizable Prompt Injection Defense via Generation of Near-Target Adversarial Examples for Alignment Training
por: Gong, Yuyang, et al.
Publicado: (2026)

TRUSTDESC: Preventing Tool Poisoning in LLM Applications via Trusted Description Generation
por: Ye, Hengkai, et al.
Publicado: (2026)

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs
por: Zhao, Shuai, et al.
Publicado: (2025)

Like Oil and Water: Group Robustness Methods and Poisoning Defenses May Be at Odds
por: Panaitescu-Liess, Michael-Andrei, et al.
Publicado: (2025)

Defense against Joint Poison and Evasion Attacks: A Case Study of DERMS
por: Abdeen, Zain ul, et al.
Publicado: (2024)

Mitigating Data Poisoning Attacks to Local Differential Privacy
por: Li, Xiaolin, et al.
Publicado: (2025)

Generate "Normal", Edit Poisoned: Branding Injection via Hint Embedding in Image Editing
por: Sun, Desen, et al.
Publicado: (2026)

Persistent Pre-Training Poisoning of LLMs
por: Zhang, Yiming, et al.
Publicado: (2024)

Poison Attacks and Adversarial Prompts Against an Informed University Virtual Assistant
por: Fernandez, Ivan A., et al.
Publicado: (2024)

Winter Soldier: Backdooring Language Models at Pre-Training with Indirect Data Poisoning
por: Bouaziz, Wassim, et al.
Publicado: (2025)

Stealthy Poisoning Attacks Bypass Defenses in Regression Settings
por: Carnerero-Cano, Javier, et al.
Publicado: (2026)

Pandora: Jailbreak GPTs by Retrieval Augmented Generation Poisoning
por: Deng, Gelei, et al.
Publicado: (2024)

MIRAGE: Misleading Retrieval-Augmented Generation via Black-box and Query-agnostic Poisoning Attacks
por: Chen, Tailun, et al.
Publicado: (2025)

Fake Resume Attacks: Data Poisoning on Online Job Platforms
por: Yamashita, Michiharu, et al.
Publicado: (2024)

Visualizing the Shadows: Unveiling Data Poisoning Behaviors in Federated Learning
por: Zhang, Xueqing, et al.
Publicado: (2024)

Securing Voice Authentication Applications Against Targeted Data Poisoning
por: Mohammadi, Alireza, et al.
Publicado: (2024)

SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
por: He, Xuanli, et al.
Publicado: (2024)