Copertina

CodePurify: Defend Backdoor Attacks on Neural Code Models via Entropy-based Purification

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Mu, Fangwen, Wang, Junjie, Yu, Zhuohao, Shi, Lin, Wang, Song, Li, Mingyang, Wang, Qing
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Cryptography and Security Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Defending Code Language Models against Backdoor Attacks with Deceptive Cross-Entropy Loss
di: Yang, Guang, et al.
Pubblicazione: (2024)

Defending against Backdoor Attack on Deep Neural Networks
di: Cheng, Hao, et al.
Pubblicazione: (2020)

Invariant Aggregator for Defending against Federated Backdoor Attacks
di: Wang, Xiaoyang, et al.
Pubblicazione: (2022)

Transferable Backdoor Attacks for Code Models via Sharpness-Aware Adversarial Perturbation
di: Chang, Shuyu, et al.
Pubblicazione: (2026)

Double Backdoored: Converting Code Large Language Model Backdoors to Traditional Malware via Adversarial Instruction Tuning Attacks
di: Hossen, Md Imran, et al.
Pubblicazione: (2024)

Your Agent Can Defend Itself against Backdoor Attacks
di: Changjiang, Li, et al.
Pubblicazione: (2025)

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

Uncovering and Aligning Anomalous Attention Heads to Defend Against NLP Backdoor Attacks
di: Jin, Haotian, et al.
Pubblicazione: (2025)

"No Matter What You Do": Purifying GNN Models via Backdoor Unlearning
di: Zhang, Jiale, et al.
Pubblicazione: (2024)

Backdoors in Code Summarizers: How Bad Is It?
di: Wang, Chenyu, et al.
Pubblicazione: (2025)

A Practical Trigger-Free Backdoor Attack on Neural Networks
di: Wang, Jiahao, et al.
Pubblicazione: (2024)

Buffer is All You Need: Defending Federated Learning against Backdoor Attacks under Non-iids via Buffering
di: Lyu, Xingyu, et al.
Pubblicazione: (2025)

AdInject: Real-World Black-Box Attacks on Web Agents via Advertising Delivery
di: Wang, Haowei, et al.
Pubblicazione: (2025)

Joint-GCG: Unified Gradient-Based Poisoning Attacks on Retrieval-Augmented Generation Systems
di: Wang, Haowei, et al.
Pubblicazione: (2025)

Exploring the Security Threats of Retriever Backdoors in Retrieval-Augmented Code Generation
di: Li, Tian, et al.
Pubblicazione: (2025)

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

Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models
di: Wan, Guangnian, et al.
Pubblicazione: (2026)

Purifying Approximate Differential Privacy with Randomized Post-processing
di: Lin, Yingyu, et al.
Pubblicazione: (2025)

Event Trojan: Asynchronous Event-based Backdoor Attacks
di: Wang, Ruofei, et al.
Pubblicazione: (2024)

Models Are Codes: Towards Measuring Malicious Code Poisoning Attacks on Pre-trained Model Hubs
di: Zhao, Jian, et al.
Pubblicazione: (2024)

Backdoor Attacks on Discrete Graph Diffusion Models
di: Wang, Jiawen, et al.
Pubblicazione: (2025)

E-SAGE: Explainability-based Defense Against Backdoor Attacks on Graph Neural Networks
di: Yuan, Dingqiang, et al.
Pubblicazione: (2024)

Practical, Generalizable and Robust Backdoor Attacks on Text-to-Image Diffusion Models
di: Dai, Haoran, et al.
Pubblicazione: (2025)

Data Extraction Attacks in Retrieval-Augmented Generation via Backdoors
di: Peng, Yuefeng, et al.
Pubblicazione: (2024)

Class-Conditional Neural Polarizer: A Lightweight and Effective Backdoor Defense by Purifying Poisoned Features
di: Zhu, Mingli, et al.
Pubblicazione: (2025)

DeCoMa: Detecting and Purifying Code Dataset Watermarks through Dual Channel Code Abstraction
di: Xiao, Yuan, et al.
Pubblicazione: (2025)

MalModel: Hiding Malicious Payload in Mobile Deep Learning Models with Black-box Backdoor Attack
di: Hua, Jiayi, et al.
Pubblicazione: (2024)

Security Attacks on LLM-based Code Completion Tools
di: Cheng, Wen, et al.
Pubblicazione: (2024)

Injection, Attack and Erasure: Revocable Backdoor Attacks via Machine Unlearning
di: Song, Baogang, et al.
Pubblicazione: (2025)

Mimicking the Familiar: Dynamic Command Generation for Information Theft Attacks in LLM Tool-Learning System
di: Jiang, Ziyou, et al.
Pubblicazione: (2025)

Defending against Backdoor Attacks via Module Switching
di: Li, Weijun, et al.
Pubblicazione: (2025)

Watch the Watcher! Backdoor Attacks on Security-Enhancing Diffusion Models
di: Li, Changjiang, et al.
Pubblicazione: (2024)

Meme Trojan: Backdoor Attacks Against Hateful Meme Detection via Cross-Modal Triggers
di: Wang, Ruofei, et al.
Pubblicazione: (2024)

Inhibitory Attacks on Backdoor-based Fingerprinting for Large Language Models
di: Fu, Hang, et al.
Pubblicazione: (2026)

Merge Hijacking: Backdoor Attacks to Model Merging of Large Language Models
di: Yuan, Zenghui, et al.
Pubblicazione: (2025)

Revisiting Training-Inference Trigger Intensity in Backdoor Attacks
di: Lin, Chenhao, et al.
Pubblicazione: (2025)

Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors
di: Li, Zi, et al.
Pubblicazione: (2026)

JBShield: Defending Large Language Models from Jailbreak Attacks through Activated Concept Analysis and Manipulation
di: Zhang, Shenyi, et al.
Pubblicazione: (2025)

Gungnir: Exploiting Stylistic Features in Images for Backdoor Attacks on Diffusion Models
di: Zhang, Lei, et al.
Pubblicazione: (2025)

One Shot Dominance: Knowledge Poisoning Attack on Retrieval-Augmented Generation Systems
di: Chang, Zhiyuan, et al.
Pubblicazione: (2025)