Image de couverture de livre

Dark Distillation: Backdooring Distilled Datasets without Accessing Raw Data

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Yang, Ziyuan, Yan, Ming, Zhang, Yi, Zhou, Joey Tianyi
Format:	Preprint
Publié:	2025
Sujets:	Cryptography and Security Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

From Compression to Accountability: Harmless Copyright Protection for Dataset Distillation
par: Liang, Yan, et autres
Publié: (2026)

Clean-Label Physical Backdoor Attacks with Data Distillation
par: Dao, Thinh, et autres
Publié: (2024)

How to Backdoor the Knowledge Distillation
par: Wu, Chen, et autres
Publié: (2025)

Robust Knowledge Distillation in Federated Learning: Counteracting Backdoor Attacks
par: Alharbi, Ebtisaam, et autres
Publié: (2025)

INK: Inheritable Natural Backdoor Attack Against Model Distillation
par: Liu, Xiaolei, et autres
Publié: (2023)

Secure Federated Data Distillation
par: Arazzi, Marco, et autres
Publié: (2025)

Pay Attention to the Triggers: Constructing Backdoors That Survive Distillation
par: De Muri, Giovanni, et autres
Publié: (2025)

Unlearning Backdoor Attacks for LLMs with Weak-to-Strong Knowledge Distillation
par: Zhao, Shuai, et autres
Publié: (2024)

Turning Black Box into White Box: Dataset Distillation Leaks
par: Chen, Huajie, et autres
Publié: (2026)

Distilling Lightweight Language Models for C/C++ Vulnerabilities
par: Wei, Zhiyuan, et autres
Publié: (2025)

CL-Attack: Textual Backdoor Attacks via Cross-Lingual Triggers
par: Zheng, Jingyi, et autres
Publié: (2024)

Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward
par: Guo, Weiyang, et autres
Publié: (2026)

BackdoorMBTI: A Backdoor Learning Multimodal Benchmark Tool Kit for Backdoor Defense Evaluation
par: Yu, Haiyang, et autres
Publié: (2024)

SNEAKDOOR: Stealthy Backdoor Attacks against Distribution Matching-based Dataset Condensation
par: Yang, He, et autres
Publié: (2026)

Backdoor Attribution: Elucidating and Controlling Backdoor in Language Models
par: Yu, Miao, et autres
Publié: (2025)

Safeguarding Text-to-Image Generative Models Against Unauthorized Knowledge Distillation
par: Gao, Yilan, et autres
Publié: (2026)

DeBackdoor: A Deductive Framework for Detecting Backdoor Attacks on Deep Models with Limited Data
par: Popovic, Dorde, et autres
Publié: (2025)

BELT: Old-School Backdoor Attacks can Evade the State-of-the-Art Defense with Backdoor Exclusivity Lifting
par: Qiu, Huming, et autres
Publié: (2023)

DemonAgent: Dynamically Encrypted Multi-Backdoor Implantation Attack on LLM-based Agent
par: Zhu, Pengyu, et autres
Publié: (2025)

BAFFLE: Hiding Backdoors in Offline Reinforcement Learning Datasets
par: Gong, Chen, et autres
Publié: (2022)

OpenPort Protocol: A Security Governance Specification for AI Agent Tool Access
par: Zhu, Genliang, et autres
Publié: (2026)

Stealthy Backdoor Attack to Real-world Models in Android Apps
par: Wei, Jiali, et autres
Publié: (2025)

You Can Backdoor Personalized Federated Learning
par: Ye, Tiandi, et autres
Publié: (2023)

DistillGuard: Evaluating Defenses Against LLM Knowledge Distillation
par: Jiang, Bo
Publié: (2026)

Your LLM Agent Can Leak Your Data: Data Exfiltration via Backdoored Tool Use
par: Zhang, Wuyang, et autres
Publié: (2026)

The Ripple Effect: On Unforeseen Complications of Backdoor Attacks
par: Zhang, Rui, et autres
Publié: (2025)

Evolutionary Trigger Detection and Lightweight Model Repair Based Backdoor Defense
par: Zhou, Qi, et autres
Publié: (2024)

MISLEADER: Defending against Model Extraction with Ensembles of Distilled Models
par: Cheng, Xueqi, et autres
Publié: (2025)

Backdoor Sentinel: Detecting and Detoxifying Backdoors in Diffusion Models via Temporal Noise Consistency
par: Wang, Bingzheng, et autres
Publié: (2026)

The Stronger the Diffusion Model, the Easier the Backdoor: Data Poisoning to Induce Copyright Breaches Without Adjusting Finetuning Pipeline
par: Wang, Haonan, et autres
Publié: (2024)

SSCL-BW: Sample-Specific Clean-Label Backdoor Watermarking for Dataset Ownership Verification
par: Wang, Yingjia, et autres
Publié: (2025)

MARS: A Malignity-Aware Backdoor Defense in Federated Learning
par: Wan, Wei, et autres
Publié: (2025)

From Poisoned to Aware: Fostering Backdoor Self-Awareness in LLMs
par: Shen, Guangyu, et autres
Publié: (2025)

On Membership Inference Attacks in Knowledge Distillation
par: Cui, Ziyao, et autres
Publié: (2025)

When Backdoors Speak: Understanding LLM Backdoor Attacks Through Model-Generated Explanations
par: Ge, Huaizhi, et autres
Publié: (2024)

CBPF: Filtering Poisoned Data Based on Composite Backdoor Attack
par: Xia, Hanfeng, et autres
Publié: (2024)

Taught Well Learned Ill: Towards Distillation-conditional Backdoor Attack
par: Chen, Yukun, et autres
Publié: (2025)

AutoBackdoor: Automating Backdoor Attacks via LLM Agents
par: Li, Yige, et autres
Publié: (2025)

Backdoor4Good: Benchmarking Beneficial Uses of Backdoors in LLMs
par: Li, Yige, et autres
Publié: (2026)

Lightweight and Fast Backdoor Model Detection
par: Yu, Yinbo, et autres
Publié: (2026)