Linux 内核提权 CVE-2026-31431(copy-fail)
CVE-2026-31431
Linux 内核提权 CVE-2026-31431(copy-fail)
2026-08-24 归档
今日共收录 798 条安全情报，包含 82 个 CVE，390 篇安全通告，以及 72 篇研究论文。
← 返回归档列表Alert for CVE-2026-35273
CVE-2026-35273
Alert for CVE-2026-35273
【安全通告】PostgreSQL 堆缓冲区溢出远程代码执行漏洞风险通告（CVE-2026-14669）
CVE-2026-14669
【安全通告】PostgreSQL 堆缓冲区溢出远程代码执行漏洞风险通告（CVE-2026-14669）
GhostTac: Manipulating Tactile Sensors without Physical Contact
触觉传感器是现代机器人系统中感知物理环境的重要组件，但其物理层安全性长期缺乏关注。本文提出 GhostTac，据作者所知，这是首个通过电磁干扰（EMI）实现非接触式操纵触觉传感器的攻击方法。GhostTac 利用触觉传感器模拟前端中常见的非
QUASAR: A Quantum-Classical Neural Network for SAR Satellite Physical-Layer Authentication
本文提出 QUASAR，据作者所知这是首个面向 X 波段合成孔径雷达（SAR）卫星物理层认证（PLA）的量子-经典混合架构。X 波段 SAR 卫星（8-12 GHz）在灾害响应、环境监测和军事侦察中至关重要，但缺乏与密码学机制正交的物理层认
Enabling Threshold Custody for the Lightning Network with Nested Threshold Multi-Signatures
该论文针对比特币闪电网络通道端点依赖单一在线密钥易受攻击的问题，提出了一种新的密码学原语——嵌套阈值多签名（nested threshold multi-signatures），用于在不改变现有协议的前提下，将闪电网络参与方阈值化。作者指出
AID-Guard: Stateful Authorization for Delegated Agent Effects
本文针对工具型AI代理（Tool-using AI agents）在将委托任务转化为服务提供方（provider）副作用（effects）时存在的授权缺口问题，提出了一种名为AID-Guard的有状态授权到副作用闭环协议。当前典型的授权机制
BackDFL: A Unified Benchmark For Backdoor Attacks and Defenses In Decentralized Federated Learning
去中心化联邦学习（DFL）通过点对点模型交换取代集中式参数服务器，旨在实现无信任依赖的协作学习。然而，这种架构转变从根本上重塑了威胁面：由于缺乏全局协调的聚合机制，DFL特别容易受到后门攻击——恶意参与者可以在保持干净任务性能的同时植入持久
Keyed Provenance Watermarking with Complementary Lattice-Based Secure Aggregation for Federated Learning
联邦学习（FL）面临多层次安全威胁，包括数据泄露、未经授权的模型重用以及恶意梯度操纵。现有防御方法通常孤立地处理这些问题，缺乏端到端的统一防护框架。本文提出一种将带密钥的上下文来源水印与基于格的可验证安全聚合相结合的新型联邦学习框架，旨在同
Faults That Fortify: CNN Adversarial Robustness via GPU Undervolting
卷积神经网络（CNN）面临对抗攻击脆弱性和训练成本高昂的双重挑战。对抗训练虽有效但计算开销巨大，尤其在向能量受限的边缘设备迁移时更为突出。本文提出在训练过程中对 GPU 进行欠压（undervolting）处理，通过降低供电电压引入随机故障
ShadowPath: Lookup-Private Credential Status Verification over Authenticated State
本文提出 ShadowPath，一种面向可验证凭证（Verifiable Credentials）的隐私保护撤销状态验证方案。在可验证凭证模型中，持有者可以出示由颁发者签名的数字声明，而无需颁发者参与每次出示过程。但撤销机制使隐私模型复杂化
Trustworthy mobile edge caching: a blockchain approach to mitigate malicious nodes and incentivize cache sharing
本文针对移动边缘缓存中恶意节点和缓存容量受限的问题，提出了一种基于区块链的信任管理机制。随着移动网络流量增长，边缘服务器上的内容缓存对降低延迟至关重要，但恶意边缘服务器可能删除或篡改缓存内容，且边缘服务器容量有限。为克服容量限制，作者引入辅
net: tap: set skb->dev before parsing virtio net header in tap_get_user_xdp()
CVE-2026-74684
In the Linux kernel, the following vulnerability has been resolved: net: tap: set skb->dev before parsing virtio net he
ALSA: usx2y: bound the hwdep mmap fault offset
CVE-2026-74641
In the Linux kernel, the following vulnerability has been resolved: ALSA: usx2y: bound the hwdep mmap fault offset snd
perf/core: Fix group leader use-after-free after sibling detach
CVE-2026-74637
In the Linux kernel, the following vulnerability has been resolved: perf/core: Fix group leader use-after-free after si
mm/huge_memory: fix huge_zero_pfn race
CVE-2026-74632
In the Linux kernel, the following vulnerability has been resolved: mm/huge_memory: fix huge_zero_pfn race Patch serie
net: smc: fix splice entry lifetime imbalance in smc_rx_splice
CVE-2026-74631
In the Linux kernel, the following vulnerability has been resolved: net: smc: fix splice entry lifetime imbalance in sm
net/x25: fix use-after-free of the socket by its timers
CVE-2026-74628
In the Linux kernel, the following vulnerability has been resolved: net/x25: fix use-after-free of the socket by its ti
NTB: ntb_netdev: Preserve RX queue depth on allocation failure
CVE-2026-74626
In the Linux kernel, the following vulnerability has been resolved: NTB: ntb_netdev: Preserve RX queue depth on allocat
netfilter: nf_conntrack: defer invalid log until after unlock
CVE-2026-74624
In the Linux kernel, the following vulnerability has been resolved: netfilter: nf_conntrack: defer invalid log until af