Private LoRA Fine-Tuning of Open-Source LLMs with Homomorphic Encryption

Preserving data confidentiality during the fine-tuning of open-source Large Language Models LLMs is crucial for sensitive applications. This work introduces an interactive protocol adapting the Low-Rank Adaptation LoRA technique for private fine-tuning. Homomorphic Encryption HE protects the...

Optimizing Mouse Dynamics for User Authentication by Machine Learning: Addressing Data Sufficiency, Accuracy-Practicality Trade-Off, and Model Performance Challenges

User authentication is essential to ensure secure access to computer systems, yet traditional methods face limitations in usability, cost, and security. Mouse dynamics authentication, based on the analysis of users' natural interaction behaviors with mouse devices, offers a cost-effective,...

SUSE CVE-2025-37870

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

Privacy-Aware Berrut Approximated Coded Computing Applied to General Distributed Learning

Coded computing is one of the techniques that can be used for privacy protection in Federated Learning. However, most of the constructions used for coded computing work only under the assumption that the computations involved are exact, generally restricted to special classes of functions, and...

CVE-2025-37870

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

AZL-62589 CVE-2025-37870 affecting package kernel 6.6.126.1-1

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

DEBIAN-CVE-2025-37870

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

UBUNTU-CVE-2025-37870

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

CVE-2025-37870 drm/amd/display: prevent hang on link training fail

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

CVE-2025-37870

Technical details about CVE-2025-37870 are not publicly provided in the supplied documents. Please monitor for updates from vendors and security advisories.

CVE-2025-37870 drm/amd/display: prevent hang on link training fail

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

CVE-2025-37870

In the Linux kernel, the following vulnerability has been resolved: drm/amd/display: prevent hang on link training fail Why When link training fails, the phy clock will be disabled. However, in enablestreams, it is assumed that link training succeeded and the mux selects the phy clock, causing a...

Learning from the Good Ones: Risk Profiling-Based Defenses against Evasion Attacks on DNNs

Safety-critical applications such as healthcare and autonomous vehicles use deep neural networks DNN to make predictions and infer decisions. DNNs are susceptible to evasion attacks, where an adversary crafts a malicious data instance to trick the DNN into making wrong decisions at inference time...

Linux kernel 安全漏洞

Linux kernel is the kernel used by Linux, the open source operating system of the Linux Foundation in the United States. A security vulnerability exists in the Linux kernel that stems from the failure to switch the clock source when amd displays a driver link training failure, which could cause t...

PT-2025-20520

Name of the Vulnerable Software and Affected Versions Linux kernel affected versions not specified Description The issue occurs in the Linux kernel when link training fails, causing the phy clock to be disabled. However, the code assumes link training succeeded, leading to a hang when a register...

Engineering Risk-Aware, Security-By-Design Frameworks for Assurance of Large-Scale Autonomous AI Models

As AI models scale to billions of parameters and operate with increasing autonomy, ensuring their safe, reliable operation demands engineering-grade security and assurance frameworks. This paper presents an enterprise-level, risk-aware, security-by-design approach for large-scale autonomous AI...

Privacy-Preserving Transformers: SwiftKey'S Differential Privacy Implementation

In this paper we train a transformer using differential privacy DP for language modeling in SwiftKey. We run multiple experiments to balance the trade-off between the model size, run-time speed and accuracy. We show that we get small and consistent gains in the next-word-prediction and accuracy...

MergeGuard: Efficient Thwarting of Trojan Attacks in Machine Learning Models

This paper proposes MergeGuard, a novel methodology for mitigation of AI Trojan attacks. Trojan attacks on AI models cause inputs embedded with triggers to be misclassified to an adversary's target class, posing a significant threat to model usability trained by an untrusted third party. The core...

BadLingual: a Novel Lingual-Backdoor Attack against Large Language Models

In this paper, we present a new form of backdoor attack against Large Language Models LLMs: lingual-backdoor attacks. The key novelty of lingual-backdoor attacks is that the language itself serves as the trigger to hijack the infected LLMs to generate inflammatory speech. They enable the precise...

LLM Security: Vulnerabilities, Attacks, Defenses, and Countermeasures

As large language models LLMs continue to evolve, it is critical to assess the security threats and vulnerabilities that may arise both during their training phase and after models have been deployed. This survey seeks to define and categorize the various attacks targeting LLMs, distinguishing...