Improving LLM Agents with Reinforcement Learning on Cryptographic CTF Challenges

Large Language Models LLMs still struggle with the structured reasoning and tool-assisted computation needed for problem solving in cybersecurity applications. In this work, we introduce "random-crypto", a cryptographic Capture-the-Flag CTF challenge generator framework that we use to fine-tune a...

Practical Adversarial Attacks on Stochastic Bandits Via Fake Data Injection

Adversarial attacks on stochastic bandits have traditionally relied on some unrealistic assumptions, such as per-round reward manipulation and unbounded perturbations, limiting their relevance to real-world systems. We propose a more practical threat model, Fake Data Injection, which reflects...

PYSEC-2025-54

vLLM is an inference and serving engine for large language models LLMs. In versions 0.8.0 up to but excluding 0.9.0, hitting the /v1/completions API with a invalid jsonschema as a Guided Param kills the vllm server. This vulnerability is similar GHSA-9hcf-v7m4-6m2j/CVE-2025-48943, but for regex...

PYSEC-2025-50

vLLM, an inference and serving engine for large language models LLMs, has a Regular Expression Denial of Service ReDoS vulnerability in the file vllm/entrypoints/openai/toolparsers/pythonictoolparser.py of versions 0.6.4 up to but excluding 0.9.0. The root cause is the use of a highly complex and...

CVE-2025-4986 Stored Cross-site Scripting (XSS) vulnerability affecting Model Definition in Product Manager from Release 3DEXPERIENCE R2022x through Release 3DEXPERIENCE R2025x

A stored Cross-site Scripting XSS vulnerability affecting Model Definition in Product Manager from Release 3DEXPERIENCE R2022x through Release 3DEXPERIENCE R2025x allows an attacker to execute arbitrary script code in user's browser session...

CVE-2025-4986

CVE-2025-4986 is a stored XSS vulnerability in Dassault Systèmes’ 3DEXPERIENCE Product Manager, specifically the Model Definition feature, affecting releases from R2022x to R2025x. The captured details indicate that an attacker can execute arbitrary script in a user’s browser session via this sto...

CVE-2025-4986 Stored Cross-site Scripting (XSS) vulnerability affecting Model Definition in Product Manager from Release 3DEXPERIENCE R2022x through Release 3DEXPERIENCE R2025x

A stored Cross-site Scripting XSS vulnerability affecting Model Definition in Product Manager from Release 3DEXPERIENCE R2022x through Release 3DEXPERIENCE R2025x allows an attacker to execute arbitrary script code in user's browser session...

Shadow Defense against Gradient Inversion Attack in Federated Learning

Federated learning FL has emerged as a transformative framework for privacy-preserving distributed training, allowing clients to collaboratively train a global model without sharing their local data. This is especially crucial in sensitive fields like healthcare, where protecting patient data is...

Hush! Protecting Secrets during Model Training: an Indistinguishability Approach

We consider the problem of secret protection, in which a business or organization wishes to train a model on their own data, while attempting to not leak secrets potentially contained in that data via the model. The standard method for training models to avoid memorization of secret information i...

vLLM 安全漏洞

vLLM is a high throughput and memory efficient inference and service engine for LLM from the vLLM open source. A security vulnerability exists in vLLM versions prior to 0.6.4 through 0.9.0 that stems from a complex regular expression used in tool call detection that could lead to a regular...

Breaking the Gold Standard: Extracting Forgotten Data under Exact Unlearning in Large Language Models

Large language models are typically trained on datasets collected from the web, which may inadvertently contain harmful or sensitive personal information. To address growing privacy concerns, unlearning methods have been proposed to remove the influence of specific data from trained models. Of...

vLLM 输入验证错误漏洞

vLLM is a high throughput and memory efficient inference and service engine for LLM from the vLLM open source. An input validation error vulnerability exists in vLLM versions prior to 0.8.0 through 0.9.0, which stems from accidental or malformed inputs in the pattern and type fields that are not...

HCL Traveler 代码问题漏洞

HCL Traveler is a software from HCL India. It is used to provide automatic, bi-directional, wireless synchronization between HCL Domino servers and wireless handheld devices. A security vulnerability exists in HCL Traveler for Microsoft Outlook that stems from vulnerability to COM hijacking attac...

Chances and Challenges of the Model Context Protocol in Digital Forensics and Incident Response

Large language models hold considerable promise for supporting forensic investigations, but their widespread adoption is hindered by a lack of transparency, explainability, and reproducibility. This paper explores how the emerging Model Context Protocol can address these challenges and support th...

LPASS: Linear Probes As Stepping Stones for Vulnerability Detection Using Compressed LLMs

Large Language Models LLMs are being extensively used for cybersecurity purposes. One of them is the detection of vulnerable codes. For the sake of efficiency and effectiveness, compression and fine-tuning techniques are being developed, respectively. However, they involve spending substantial...

WordPress The Fashion - Model Agency One Page Beauty Theme plugin <= 1.4.4 - Deserialization of untrusted data Vulnerability

WordPress The Fashion - Model Agency One Page Beauty Theme plugin = 1.4.4 - Deserialization of untrusted data Vulnerability discovered by Tran Nguyen Bao Khanh VCI - VNPT Cyber Immunity in WordPress Theme The Fashion - Model Agency One Page Beauty Theme versions = 1.4.4...

MINI-QQX2-38MR-RH26

Bulletin has no description...

MINI-HRVR-C39M-VC8R

Bulletin has no description...

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention

Cautious predictions -- where a machine learning model abstains when uncertain -- are crucial for limiting harmful errors in safety-critical applications. In this work, we identify a novel threat: a dishonest institution can exploit these mechanisms to discriminate or unjustly deny services under...

Hijacking Large Language Models Via Adversarial In-Context Learning

In-context learning ICL has emerged as a powerful paradigm leveraging LLMs for specific downstream tasks by utilizing labeled examples as demonstrations demos in the preconditioned prompts. Despite its promising performance, crafted adversarial attacks pose a notable threat to the robustness of...