American Fuzzy Lop plus plus 4.34c

Google's American Fuzzy Lop is a brute-force fuzzer coupled with an exceedingly simple but rock-solid instrumentation-guided genetic algorithm. afl++ is a superior fork to Google's afl. It has more speed, more and better mutations, more and better instrumentation, custom module support, etc...

SecInfer: Preventing Prompt Injection Via Inference-Time Scaling

Prompt injection attacks pose a pervasive threat to the security of Large Language Models LLMs. State-of-the-art prevention-based defenses typically rely on fine-tuning an LLM to enhance its security, but they achieve limited effectiveness against strong attacks. In this work, we propose...

Guided Reasoning in LLM-Driven Penetration Testing Using Structured Attack Trees

Recent advances in Large Language Models LLMs have driven interest in automating cybersecurity penetration testing workflows, offering the promise of faster and more consistent vulnerability assessment for enterprise systems. Existing LLM agents for penetration testing primarily rely on self-guid...

Unlearning at Scale: Implementing the Right to Be Forgotten in Large Language Models

We study the right to be forgotten GDPR Art. 17 for large language models and frame unlearning as a reproducible systems problem. Our approach treats training as a deterministic program and logs a minimal per-microbatch record ordered ID hash, RNG seed, learning-rate value, optimizer-step counter...

BlindGuard: Safeguarding LLM-Based Multi-Agent Systems under Unknown Attacks

The security of LLM-based multi-agent systems MAS is critically threatened by propagation vulnerability, where malicious agents can distort collective decision-making through inter-agent message interactions. While existing supervised defense methods demonstrate promising performance, they may be...

When LLMs Copy to Think: Uncovering Copy-Guided Attacks in Reasoning LLMs

Large Language Models LLMs have become integral to automated code analysis, enabling tasks such as vulnerability detection and code comprehension. However, their integration introduces novel attack surfaces. In this paper, we identify and investigate a new class of prompt-based attacks, termed...

Human-Centered Interactive Anonymization for Privacy-Preserving Machine Learning: a Case for Human-Guided K-Anonymity

Privacy-preserving machine learning ML seeks to balance data utility and privacy, especially as regulations like the GDPR mandate the anonymization of personal data for ML applications. Conventional anonymization approaches often reduce data utility due to indiscriminate generalization or...

SecureT2I: No More Unauthorized Manipulation on AI Generated Images from Prompts

Text-guided image manipulation with diffusion models enables flexible and precise editing based on prompts, but raises ethical and copyright concerns due to potential unauthorized modifications. To address this, we propose SecureT2I, a secure framework designed to prevent unauthorized editing in...

American Fuzzy Lop plus plus 4.33c

Google's American Fuzzy Lop is a brute-force fuzzer coupled with an exceedingly simple but rock-solid instrumentation-guided genetic algorithm. afl++ is a superior fork to Google's afl. It has more speed, more and better mutations, more and better instrumentation, custom module support, etc...

KnowML: Improving Generalization of ML-NIDS with Attack Knowledge Graphs

Despite extensive research on Machine Learning-based Network Intrusion Detection Systems ML-NIDS, their capability to detect diverse attack variants remains uncertain. Prior studies have largely relied on homogeneous datasets, which artificially inflate performance scores and offer a false sense ...

InverTune: Removing Backdoors from Multimodal Contrastive Learning Models Via Trigger Inversion and Activation Tuning

Multimodal contrastive learning models like CLIP have demonstrated remarkable vision-language alignment capabilities, yet their vulnerability to backdoor attacks poses critical security risks. Attackers can implant latent triggers that persist through downstream tasks, enabling malicious control ...

ReGA: Representation-Guided Abstraction for Model-Based Safeguarding of LLMs

Large Language Models LLMs have achieved significant success in various tasks, yet concerns about their safety and security have emerged. In particular, they pose risks in generating harmful content and vulnerability to jailbreaking attacks. To analyze and monitor machine learning models,...

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...

Uncaught Exception

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Uncaught Exception through the guidedregex parameter when using xgrammar validation. An attacker can cause the application to crash by sending an...

Uncaught Exception

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Uncaught Exception through the guidedjson parameter when using xgrammar validation. An attacker can cause the application to crash by sending an...

PT-2025-23226 · Vllm · Vllm

Name of the Vulnerable Software and Affected Versions: vLLM versions 0.8.0 through 0.9.0 Description: The issue arises when the /v1/completions API endpoint is hit with an invalid json schema as a Guided Param, causing the vLLM server to crash. This is similar to a previously known issue but...

CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models

Retrieval-Augmented Generation RAG enhances large language models LLMs by incorporating external knowledge, but its openness introduces vulnerabilities that can be exploited by poisoning attacks. Existing poisoning methods for RAG systems have limitations, such as poor generalization and lack of...

Fixing 7,400 Bugs for 1$: Cheap Crash-Site Program Repair

The rapid advancement of bug-finding techniques has led to the discovery of more vulnerabilities than developers can reasonably fix, creating an urgent need for effective Automated Program Repair APR methods. However, the complexity of modern bugs often makes precise root cause analysis difficult...

CVE-2023-22029

Vulnerability in the Oracle Commerce Guided Search product of Oracle Commerce component: Workbench. The supported version that is affected is 11.3.2. Easily exploitable vulnerability allows unauthenticated attacker with network access via HTTP to compromise Oracle Commerce Guided Search. Successf...

CVE-2021-2345

Vulnerability in the Oracle Commerce Guided Search / Oracle Commerce Experience Manager product of Oracle Commerce component: Tools and Frameworks. The supported version that is affected is 11.3.1.5. Easily exploitable vulnerability allows low privileged attacker with network access via HTTP to...