Backdoor Attacks against Patch-Based Mixture of Experts

As Deep Neural Networks DNNs continue to require larger amounts of data and computational power, Mixture of Experts MoE models have become a popular choice to reduce computational complexity. This popularity increases the importance of considering the security of MoE architectures. Unfortunately,...

A Survey on Privacy Risks and Protection in Large Language Models

Although Large Language Models LLMs have become increasingly integral to diverse applications, their capabilities raise significant privacy concerns. This survey offers a comprehensive overview of privacy risks associated with LLMs and examines current solutions to mitigate these challenges. Firs...

Protocol-Agnostic and Data-Free Backdoor Attacks on Pre-Trained Models in RF Fingerprinting

While supervised deep neural networks DNNs have proven effective for device authentication via radio frequency RF fingerprinting, they are hindered by domain shift issues and the scarcity of labeled data. The success of large language models has led to increased interest in unsupervised pre-train...

Malicious code in helmet-fastapi (PyPI)

--- -= Per source details. Do not edit below this line.=- Source: kam193 c1f805932ecbcd95197e98c6e2336eb773252abf5615fe135076d1848cb90395 Package contains hidden code adding a backdoor - a WebSocket path handler which will execute commands sent by an attacker knowing the path. In addition, it add...

MAL-2025-191752 Malicious code in helmet-fastapi (PyPI)

--- -= Per source details. Do not edit below this line.=- Source: kam193 c1f805932ecbcd95197e98c6e2336eb773252abf5615fe135076d1848cb90395 Package contains hidden code adding a backdoor - a WebSocket path handler which will execute commands sent by an attacker knowing the path. In addition, it add...

Sneaky WordPress Malware Disguised as Anti-Malware Plugin

WordPress sites are under threat from a deceptive anti-malware plugin. Learn how this malware grants backdoor access, hides…...

How to Backdoor the Knowledge Distillation

Knowledge distillation has become a cornerstone in modern machine learning systems, celebrated for its ability to transfer knowledge from a large, complex teacher model to a more efficient student model. Traditionally, this process is regarded as secure, assuming the teacher model is clean. This...

Cert-SSB: toward Certified Sample-Specific Backdoor Defense

Deep neural networks DNNs are vulnerable to backdoor attacks, where an attacker manipulates a small portion of the training data to implant hidden backdoors into the model. The compromised model behaves normally on clean samples but misclassifies backdoored samples into the attacker-specified...

SentinelOne Uncovers Chinese Espionage Campaign Targeting Its Infrastructure and Clients

Cybersecurity company SentinelOne has revealed that a China-nexus threat cluster dubbed PurpleHaze conducted reconnaissance attempts against its infrastructure and some of its high-value customers. "We first became aware of this threat cluster during a 2024 intrusion conducted against an...

SFIBA: Spatial-Based Full-Target Invisible Backdoor Attacks

Multi-target backdoor attacks pose significant security threats to deep neural networks, as they can preset multiple target classes through a single backdoor injection. This allows attackers to control the model to misclassify poisoned samples with triggers into any desired target class during...

Erased but Not Forgotten: How Backdoors Compromise Concept Erasure

The expansion of large-scale text-to-image diffusion models has raised growing concerns about their potential to generate undesirable or harmful content, ranging from fabricated depictions of public figures to sexually explicit images. To mitigate these risks, prior work has devised machine...

FFCBA: Feature-Based Full-Target Clean-Label Backdoor Attacks

Backdoor attacks pose a significant threat to deep neural networks, as backdoored models would misclassify poisoned samples with specific triggers into target classes while maintaining normal performance on clean samples. Among these, multi-target backdoor attacks can simultaneously target multip...

BadMoE: Backdooring Mixture-Of-Experts LLMs Via Optimizing Routing Triggers and Infecting Dormant Experts

Mixture-of-Experts MoE have emerged as a powerful architecture for large language models LLMs, enabling efficient scaling of model capacity while maintaining manageable computational costs. The key advantage lies in their ability to route different tokens to different "expert'' networks within th...

Russian organizations targeted by backdoor masquerading as secure networking software updates

As we were looking into a cyberincident in April 2025, we uncovered a rather sophisticated backdoor. It targeted various large organizations in Russia, spanning the government, finance, and industrial sectors. While our investigation into the attack associated with the backdoor is still ongoing, ...

Lotus Panda Hacks SE Asian Governments With Browser Stealers and Sideloaded Malware

The China-linked cyber espionage group tracked as Lotus Panda has been attributed to a campaign that compromised multiple organizations in an unnamed Southeast Asian country between August 2024 and February 2025. "Targets included a government ministry, an air traffic control organization, a...

TrojanDam: Detection-Free Backdoor Defense in Federated Learning through Proactive Model Robustification Utilizing OOD Data

Federated learning FL systems allow decentralized data-owning clients to jointly train a global model through uploading their locally trained updates to a centralized server. The property of decentralization enables adversaries to craft carefully designed backdoor updates to make the global model...

Backdoor Defense in Diffusion Models Via Spatial Attention Unlearning

Text-to-image diffusion models are increasingly vulnerable to backdoor attacks, where malicious modifications to the training data cause the model to generate unintended outputs when specific triggers are present. While classification models have seen extensive development of defense mechanisms,...

Gr33n Radar Backdoor 0.1

Gr33n Radar Backdoor is a PHP web shell backdoor that has some innovative features not normally found in web shells...

BadApex: Backdoor Attack Based on Adaptive Optimization Mechanism of Black-Box Large Language Models

Previous insertion-based and paraphrase-based backdoors have achieved great success in attack efficacy, but they ignore the text quality and semantic consistency between poisoned and clean texts. Although recent studies introduce LLMs to generate poisoned texts and improve the stealthiness,...

REDEditing: Relationship-Driven Precise Backdoor Poisoning on Text-To-Image Diffusion Models

The rapid advancement of generative AI highlights the importance of text-to-image T2I security, particularly with the threat of backdoor poisoning. Timely disclosure and mitigation of security vulnerabilities in T2I models are crucial for ensuring the safe deployment of generative models. We...