SecFwT: Efficient Privacy-Preserving Fine-Tuning of Large Language Models Using Forward-Only Passes

Large language models LLMs have transformed numerous fields, yet their adaptation to specialized tasks in privacy-sensitive domains, such as healthcare and finance, is constrained by the scarcity of accessible training data due to stringent privacy requirements. Secure multi-party computation...

CVE-2025-34021 Selea Targa IP OCR-ANPR Camera Server-Side Request Forgery

A server-side request forgery SSRF vulnerability exists in multiple Selea Targa IP OCR-ANPR camera models, including iZero, Targa 512, Targa 504, Targa Semplice, Targa 704 TKM, Targa 805, Targa 710 INOX, Targa 750, and Targa 704 ILB. The application fails to validate user-supplied input in JSON...

CVE-2025-48059

PowSyBl Core contains a polynomial Regular Expression Denial of Service (ReDoS) in the RegexCriterion class used by powsybl-iidm-criteria (versions 6.3.0–6.7.1 and powsybl-contingency-api 5.0.0–6.3.0). The vulnerability arises from unvalidated user-supplied regex patterns compiled and evaluated a...

Global Microprocessor Correctness in the Presence of Transient Execution

Correctness for microprocessors is generally understood to be conformance with the associated instruction set architecture ISA. This is the basis for one of the most important abstractions in computer science, allowing hardware designers to develop highly-optimized processors that are functionall...

From Thinking to Output: Chain-Of-Thought and Text Generation Characteristics in Reasoning Language Models

Recently, there have been notable advancements in large language models LLMs, demonstrating their growing abilities in complex reasoning. However, existing research largely overlooks a thorough and systematic comparison of these models' reasoning processes and outputs, particularly regarding thei...

H3C多款产品 安全漏洞

H3C ER2200G2 and others are products of China's Xinhua San H3C.H3C ER2200G2 is an enterprise router.H3C ERG2-450W is a wireless router.H3C ERG2-1200W is a wireless router. A security vulnerability exists in various H3C products that stems from authentication bypass and could lead to remote comman...

A Common Pool of Privacy Problems: Legal and Technical Lessons from a Large-Scale Web-Scraped Machine Learning Dataset

We investigate the contents of web-scraped data for training AI systems, at sizes where human dataset curators and compilers no longer manually annotate every sample. Building off of prior privacy concerns in machine learning models, we ask: What are the legal privacy implications of web-scraped...

Towards Effective Complementary Security Analysis Using Large Language Models

A key challenge in security analysis is the manual evaluation of potential security weaknesses generated by static application security testing SAST tools. Numerous false positives FPs in these reports reduce the effectiveness of security analysis. We propose using Large Language Models LLMs to...

MM-AttacKG: a Multimodal Approach to Attack Graph Construction with Large Language Models

Cyber Threat Intelligence CTI parsing aims to extract key threat information from massive data, transform it into actionable intelligence, enhance threat detection and defense efficiency, including attack graph construction, intelligence fusion and indicator extraction. Among these research topic...

Selea多款产品 安全漏洞

Selea Targa iZero and others are an optical character recognition camera for automatic license plate recognition from Selea, Italy. A security vulnerability exists in several Selea products, which stems from the /common/getfile.php script that does not validate the file parameter, potentially...

SmartGuard: Leveraging Large Language Models for Network Attack Detection through Audit Log Analysis and Summarization

End-point monitoring solutions are widely deployed in today's enterprise environments to support advanced attack detection and investigation. These monitors continuously record system-level activities as audit logs and provide deep visibility into security events. Unfortunately, existing methods ...

Five Uncomfortable Truths About LLMs in Production

Many tech professionals see integrating large language models LLMs as a simple process -just connect an API and let it run. At Wallarm, our experience has proved otherwise. Through rigorous testing and iteration, our engineering team uncovered several critical insights about deploying LLMs secure...

Differentiation-Based Extraction of Proprietary Data from Fine-Tuned LLMs

The increasing demand for domain-specific and human-aligned Large Language Models LLMs has led to the widespread adoption of Supervised Fine-Tuning SFT techniques. SFT datasets often comprise valuable instruction-response pairs, making them highly valuable targets for potential extraction. This...

The Hitchhiker'S Guide to Efficient, End-To-End, and Tight DP Auditing

This paper systematizes research on auditing Differential Privacy DP techniques, aiming to identify key insights into the current state of the art and open challenges. First, we introduce a comprehensive framework for reviewing work in the field and establish three cross-contextual desiderata tha...

SecureFed: a Two-Phase Framework for Detecting Malicious Clients in Federated Learning

Federated Learning FL protects data privacy while providing a decentralized method for training models. However, because of the distributed schema, it is susceptible to adversarial clients that could alter results or sabotage model performance. This study presents SecureFed, a two-phase FL...

Exploring Traffic Simulation and Cybersecurity Strategies Using Large Language Models

Intelligent Transportation Systems ITS are increasingly vulnerable to sophisticated cyberattacks due to their complex, interconnected nature. Ensuring the cybersecurity of these systems is paramount to maintaining road safety and minimizing traffic disruptions. This study presents a novel...

WormGPT Makes a Comeback Using Jailbroken Grok and Mixtral Models

Cato CTRL uncovers new WormGPT variants on Telegram powered by jailbroken Grok and Mixtral. Learn how cybercriminals jailbreak top LLMs for uncensored, illegal activities in this latest threat research...

A Nested Watermark for Large Language Models

The rapid advancement of large language models LLMs has raised concerns regarding their potential misuse, particularly in generating fake news and misinformation. To address these risks, watermarking techniques for autoregressive language models have emerged as a promising means for detecting...

Dynamic Risk Assessments for Offensive Cybersecurity Agents

Foundation models are increasingly becoming better autonomous programmers, raising the prospect that they could also automate dangerous offensive cyber-operations. Current frontier model audits probe the cybersecurity risks of such agents, but most fail to account for the degrees of freedom...

TP-Link Router Flaw CVE-2023-33538 Under Active Exploit, CISA Issues Immediate Alert

The U.S. Cybersecurity and Infrastructure Security Agency CISA on Monday added a high-severity security flaw in TP-Link wireless routers to its Known Exploited Vulnerabilities KEV catalog, citing evidence of active exploitation. The vulnerability in question is CVE-2023-33538 CVSS score: 8.8, a...