Everything You Wanted to Know about LLM-Based Vulnerability Detection but Were Afraid to Ask

Large Language Models are a promising tool for automated vulnerability detection, thanks to their success in code generation and repair. However, despite widespread adoption, a critical question remains: Are LLMs truly effective at detecting real-world vulnerabilities? Current evaluations, which...

Research Briefing: MCP Security

The present and future of security for the Model Context Protocol...

GraphAttack: Exploiting Representational Blindspots in LLM Safety Mechanisms

Large Language Models LLMs have been equipped with safety mechanisms to prevent harmful outputs, but these guardrails can often be bypassed through "jailbreak" prompts. This paper introduces a novel graph-based approach to systematically generate jailbreak prompts through semantic transformations...

Leveraging Functional Encryption and Deep Learning for Privacy-Preserving Traffic Forecasting

Over the past few years, traffic congestion has continuously plagued the nation's transportation system creating several negative impacts including longer travel times, increased pollution rates, and higher collision risks. To overcome these challenges, Intelligent Transportation Systems ITS aim ...

Adversary-Augmented Simulation for Fairness Evaluation and Defense in Hyperledger Fabric

This paper presents an adversary model and a simulation framework specifically tailored for analyzing attacks on distributed systems composed of multiple distributed protocols, with a focus on assessing the security of blockchain networks. Our model classifies and constrains adversarial actions...

DEBIAN-CVE-2025-22872

The tokenizer incorrectly interprets tags with unquoted attribute values that end with a solidus character / as self-closing. When directly using Tokenizer, this can result in such tags incorrectly being marked as self-closing, and when using the Parse functions, this can result in content...

AZL-60545 CVE-2025-22872 affecting package cf-cli for versions less than 8.7.11-3

The tokenizer incorrectly interprets tags with unquoted attribute values that end with a solidus character / as self-closing. When directly using Tokenizer, this can result in such tags incorrectly being marked as self-closing, and when using the Parse functions, this can result in content...

AZL-60450 CVE-2025-22872 affecting package keda for versions less than 2.14.1-7

The tokenizer incorrectly interprets tags with unquoted attribute values that end with a solidus character / as self-closing. When directly using Tokenizer, this can result in such tags incorrectly being marked as self-closing, and when using the Parse functions, this can result in content...

AZL-61750 CVE-2025-22872 affecting package yq 4.45.1-1

The tokenizer incorrectly interprets tags with unquoted attribute values that end with a solidus character / as self-closing. When directly using Tokenizer, this can result in such tags incorrectly being marked as self-closing, and when using the Parse functions, this can result in content...

AZL-60523 CVE-2025-22872 affecting package sriov-network-device-plugin for versions less than 3.7.0-4

The tokenizer incorrectly interprets tags with unquoted attribute values that end with a solidus character / as self-closing. When directly using Tokenizer, this can result in such tags incorrectly being marked as self-closing, and when using the Parse functions, this can result in content...

AZL-60586 CVE-2025-22872 affecting package cri-tools for versions less than 1.29.0-8

The tokenizer incorrectly interprets tags with unquoted attribute values that end with a solidus character / as self-closing. When directly using Tokenizer, this can result in such tags incorrectly being marked as self-closing, and when using the Parse functions, this can result in content...

The Digital Cybersecurity Expert: How Far Have We Come?

The increasing deployment of large language models LLMs in the cybersecurity domain underscores the need for effective model selection and evaluation. However, traditional evaluation methods often overlook specific cybersecurity knowledge gaps that contribute to performance limitations. To addres...

Mattermost 安全漏洞

Mattermost is an open source collaboration platform from Mattermost, Inc. in the United States. Mattermost suffers from an information disclosure vulnerability. The vulnerability stems from an under-restricted LLM request domain. An attacker can exploit the vulnerability to perform prompt injecti...

Provable Secure Steganography Based on Adaptive Dynamic Sampling

The security of private communication is increasingly at risk due to widespread surveillance. Steganography, a technique for embedding secret messages within innocuous carriers, enables covert communication over monitored channels. Provably Secure Steganography PSS is state of the art for making...

OpDiffer: LLM-Assisted Opcode-Level Differential Testing of Ethereum Virtual Machine

As Ethereum continues to thrive, the Ethereum Virtual Machine EVM has become the cornerstone powering tens of millions of active smart contracts. Intuitively, security issues in EVMs could lead to inconsistent behaviors among smart contracts or even denial-of-service of the entire blockchain...

FastChat 代码问题漏洞

FastChat is an open source platform from LMSYS for training, deploying and evaluating chatbots based on large language models. A code issue vulnerability exists in FastChat version 0.2.36 and earlier, which stems from a deserialization issue in the splitfiles/applydeltalowcpumem function in the...

CVE-2025-3622 Xorbits Inference model.py load deserialization

A vulnerability, which was classified as critical, has been found in Xorbits Inference up to 1.4.1. This issue affects the function load of the file xinference/thirdparty/cosyvoice/cli/model.py. The manipulation leads to deserialization...

Security Bulletin: IBM Controller is affected by vulnerabilities

Summary There are vulnerabilities in IBM® Websphere Application Server Liberty and Open-Source Software OSS components used by IBM Controller. Additionally, due to weak password requirements, IBM Controller is susceptible to compromised user accounts. Please refer to the table in the Related...

MULTI-LF: a Unified Continuous Learning Framework for Real-Time DDoS Detection in Multi-Environment Networks

Detecting Distributed Denial of Service DDoS attacks in Multi-Environment M-En networks presents significant challenges due to diverse malicious traffic patterns and the evolving nature of cyber threats. Existing AI-based detection systems struggle to adapt to new attack strategies and lack...

How to Enhance Downstream Adversarial Robustness (Almost) without Touching the Pre-Trained Foundation Model?

With the rise of powerful foundation models, a pre-training-fine-tuning paradigm becomes increasingly popular these days: A foundation model is pre-trained using a huge amount of data from various sources, and then the downstream users only need to fine-tune and adapt it to specific downstream...