Linux kernel 安全漏洞

The Linux kernel is the core of the open-source operating system Linux, developed by the Linux Foundation in the United States. There is a security vulnerability in the Linux kernel, which stems from failing to ensure sufficient space for the skb header before calling skbpush. This vulnerability...

Sparse Autoencoders Are Capable LLM Jailbreak Mitigators

Jailbreak attacks remain a persistent threat to large language model safety. We propose Context-Conditioned Delta Steering CC-Delta, an SAE-based defense that identifies jailbreak-relevant sparse features by comparing token-level representations of the same harmful request with and without...

CVE-2026-1837

A specially-crafted file can cause libjxl's decoder to write pixel data to uninitialized unallocated memory. Soon after that data from another uninitialized unallocated region is copied to pixel data. This can be done by requesting color transformation of grayscale images to another grayscale col...

[SECURITY] Fedora 42 Update: rust-scx_layered-0.0.6-7.fc42

A highly configurable multi-layer BPF / user space hybrid scheduler used within schedext, which is a Linux kernel feature which enables implementing kernel thread schedulers in BPF and dynamically loading them. https://github.com/sched-ext/scx/tree/main...

[SECURITY] Fedora 42 Update: rust-scx_rusty-0.5.4-7.fc42

A multi-domain, BPF / user space hybrid scheduler used within schedext, which is a Linux kernel feature which enables implementing kernel thread schedulers in BPF and dynamically loading them. https://github.com/sched-ext/scx/tree/main...

[SECURITY] Fedora 42 Update: rust-scx_rustland-0.0.3-7.fc42

A BPF component dispatcher that implements the low level sched-ext functionalities and a user-space counterpart scheduler, written in Rust, that implements the actual scheduling policy. This is used within schedext, which is a Linux kernel feature which enables implementing kernel thread schedule...

[SECURITY] Fedora 42 Update: open-vm-tools-13.0.10-2.fc42

The open-vm-tools project is an open source implementation of VMware Tools. It is a suite of open source virtualization utilities and drivers to improve the functionality, user experience and administration of VMware virtual machines. This package contains only the core user-space programs and...

Kill It with FIRE: On Leveraging Latent Space Directions for Runtime Backdoor Mitigation in Deep Neural Networks

Machine learning models are increasingly present in our everyday lives; as a result, they become targets of adversarial attackers seeking to manipulate the systems we interact with. A well-known vulnerability is a backdoor introduced into a neural network by poisoned training data or a malicious...

CVE-2026-26006

AutoGPT (significant-gravitas/autogpt) before version 0.6.32 is vulnerable in the Code Extraction Block due to two adjacent quantifiers in regex patterns that can cause catastrophic backtracking with long sequences of spaces, leading to DoS. The fix is to upgrade to 0.6.32. If upgrading is not po...

CVE-2026-26006 Redos (Regular Expression Denial of Service) at Code Extraction Block in significant-gravitas/autogpt

AutoGPT is a platform that allows users to create, deploy, and manage continuous artificial intelligence agents that automate complex workflows. The autogpt before 0.6.32 is vulnerable to Regular Expression Denial of Service due to the use of regex at Code Extraction Block. The two Regex are used...

CVE-2025-48517

Insufficient Granularity of Access Control in SEV firmware could allow a privileged user with a malicious hypervisor to create a SEV-ES guest with an ASID in the range meant for SEV-SNP guests potentially resulting in a partial loss of confidentiality...

UBUNTU-CVE-2025-48517

Insufficient Granularity of Access Control in SEV firmware could allow a privileged user with a malicious hypervisor to create a SEV-ES guest with an ASID in the range meant for SEV-SNP guests potentially resulting in a partial loss of confidentiality...

CVE-2025-48517

Insufficient Granularity of Access Control in SEV firmware could allow a privileged user with a malicious hypervisor to create a SEV-ES guest with an ASID in the range meant for SEV-SNP guests potentially resulting in a partial loss of confidentiality...

CVE-2025-48517

Insufficient Granularity of Access Control in SEV firmware could allow a privileged user with a malicious hypervisor to create a SEV-ES guest with an ASID in the range meant for SEV-SNP guests potentially resulting in a partial loss of confidentiality...

CVE-2025-48517

CVE-2025-48517 concerns AMD SEV firmware where insufficient granularity of access control could let a privileged attacker with a malicious hypervisor create a SEV-ES guest using an ASID reserved for SEV-SNP guests, potentially causing partial confidentiality loss. Reported across multiple sources...

CVE-2025-48517

Insufficient Granularity of Access Control in SEV firmware could allow a privileged user with a malicious hypervisor to create a SEV-ES guest with an ASID in the range meant for SEV-SNP guests potentially resulting in a partial loss of confidentiality...

[SECURITY] Fedora 43 Update: rust-scx_layered-0.0.6-7.fc43

A highly configurable multi-layer BPF / user space hybrid scheduler used within schedext, which is a Linux kernel feature which enables implementing kernel thread schedulers in BPF and dynamically loading them. https://github.com/sched-ext/scx/tree/main...

Rethinking Security of Diffusion-Based Generative Steganography

Generative image steganography is a technique that conceals secret messages within generated images, without relying on pre-existing cover images. Recently, a number of diffusion model-based generative image steganography DM-GIS methods have been introduced, which effectively combat traditional...

PT-2026-7472

Name of the Vulnerable Software and Affected Versions AutoGPT versions prior to 0.6.32 Description AutoGPT is a platform for creating, deploying, and managing continuous artificial intelligence agents that automate complex workflows. Versions of AutoGPT before 0.6.32 contain a Regular Expression...

Exploring Semantic Labeling Strategies for Third-Party Cybersecurity Risk Assessment Questionnaires

Third-Party Risk Assessment TPRA is a core cybersecurity practice for evaluating suppliers against standards such as ISO/IEC 27001 and NIST. TPRA questionnaires are typically drawn from large repositories of security and compliance questions, yet tailoring assessments to organizational needs...