EUVD-2026-18522

vLLM: Processing differential in multi-channel audio downmixing enables hidden-input/moderation bypass for audio models...

GHSA-6C4R-FMH3-7RH8 vLLM: Processing differential in multi-channel audio downmixing enables hidden-input/moderation bypass for audio models

Issue Description Librosa defaults to using numpy.mean for mono downmixing tomono, while the international standard ITU-R BS.775-4 specifies a weighted downmixing algorithm. This discrepancy results in: - Inconsistency between audio heard by humans e.g., through headphones/regular speakers and...

PYSEC-2026-2803 Out-of-bounds Read in OpenCV

OpenCV 4.1.1 has an out-of-bounds read in halbaseline::vload in core/hal/intrinsse.hpp when called from computeSSDMeanNorm in modules/video/src/disflow.cpp...

PYSEC-2026-2842 Out-of-bounds Read in OpenCV

OpenCV 4.1.1 has an out-of-bounds read in halbaseline::vload in core/hal/intrinsse.hpp when called from computeSSDMeanNorm in modules/video/src/disflow.cpp...

BIT-PARSE-2026-47248 Parse Server: GraphQL "Did you mean" validation suggestions disclose schema to unauthenticated callers

Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to versions 8.6.78 and 9.9.1, Parse Server's GraphQL endpoint discloses schema metadata to unauthenticated callers through Did you mean ...? suggestions embedded in GraphQL validation-err...

CVE-2026-47248 Parse Server: GraphQL "Did you mean" validation suggestions disclose schema to unauthenticated callers

Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to versions 8.6.78 and 9.9.1-alpha.2, Parse Server's GraphQL endpoint discloses schema metadata to unauthenticated callers through Did you mean ...? suggestions embedded in GraphQL...

CVE-2026-47248 Parse Server: GraphQL "Did you mean" validation suggestions disclose schema to unauthenticated callers

Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to versions 8.6.78 and 9.9.1-alpha.2, Parse Server's GraphQL endpoint discloses schema metadata to unauthenticated callers through Did you mean ...? suggestions embedded in GraphQL...

CVE-2026-47248 Parse Server: GraphQL "Did you mean" validation suggestions disclose schema to unauthenticated callers

Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to versions 8.6.78 and 9.9.1-alpha.2, Parse Server's GraphQL endpoint discloses schema metadata to unauthenticated callers through Did you mean ...? suggestions embedded in GraphQL...

CVE-2026-47248

CVE-2026-47248 – Parse Server GraphQL schema disclosure via Did you mean …? validation messages What is affected: Parse Server (Node.js) GraphQL endpoint exposes schema metadata to unauthenticated callers through Did you mean …? suggestions embedded in GraphQL validation errors. Root cause: Valid...

NCMD: Benign-Anchored Feature Selection for Imbalanced Network Intrusion Detection

Feature selection is critical for network intrusion detection systems NIDS operating under high-dimensional, highly imbalanced traffic, as found in operational and defense networks. Traditional filter methods rank features using global statistics computed symmetrically across classes and thus fai...

GHSA-8CPH-RGR4-G5VJ Parse Server's GraphQL "Did you mean ...?" validation suggestions disclose schema to unauthenticated callers

Impact Parse Server's GraphQL endpoint discloses schema metadata to unauthenticated callers through Did you mean ...? suggestions embedded in GraphQL validation-error messages. An unauthenticated caller who knows only the public application id can iteratively send malformed queries to reconstruct...

PT-2026-45045

Name of the Vulnerable Software and Affected Versions Parse Server versions prior to 8.6.78 Parse Server versions prior to 9.9.1-alpha.2 Description The GraphQL endpoint discloses schema metadata to unauthenticated callers via "Did you mean ...?" suggestions within GraphQL validation-error...

Modernizing User Privacy Preference Measurement through GPPI: A GDPR-Aligned Privacy Preference Item Bank

Privacy measurement instruments e.g., CFIP, IUIPC, PAQ predate GDPR by over a decade and measure privacy concerns, distinct from preferences for regulatory protections e.g., data portability, erasure, automated decision-making rights. This leaves practitioners without tools to assess whether user...

AoI-Guided Client Selection for Robust and Timely Federated Intrusion Detection in Cloud-Edge Security Analytics

Federated learning FL is attractive for cloud-edge intrusion detection because it enables collaborative training over distributed telemetry without centralizing raw logs. In production security analytics pipelines, however, only a subset of clients participates in each round, and heterogeneous...

FixV2W: Correcting Invalid CVE-CWE Mappings with Knowledge Graph Embeddings

Accurate mapping between Common Vulnerabilities and Exposures CVE and Common Weakness Enumeration CWE entries is critical for effective vulnerability management and risk assessment. However, public databases, such as the National Vulnerability Database NVD, suffer from inconsistent and incomplete...

CVE-2026-34760

Summary: CVE-2026-34760 concerns vLLM’s audio processing path via Librosa. From version 0.5.5 up to before 0.18.0, Librosa used numpy.mean for mono downmix (to_mono), while ITU-R BS.775-4 specifies a weighted downmix. This mismatch creates inconsistency between audio perceived by humans and audio...

CVE-2026-34760

vLLM is an inference and serving engine for large language models LLMs. From version 0.5.5 to before version 0.18.0, Librosa defaults to using numpy.mean for mono downmixing tomono, while the international standard ITU-R BS.775-4 specifies a weighted downmixing algorithm. This discrepancy results...

CVE-2026-34760 vLLM: Downmix Implementation Differences as Attack Vectors Against Audio AI Models

vLLM is an inference and serving engine for large language models LLMs. From version 0.5.5 to before version 0.18.0, Librosa defaults to using numpy.mean for mono downmixing tomono, while the international standard ITU-R BS.775-4 specifies a weighted downmixing algorithm. This discrepancy results...

VoiceSHIELD-Small: Real-Time Malicious Speech Detection and Transcription

Voice interfaces are quickly becoming a common way for people to interact with AI systems. This also brings new security risks, such as prompt injection, social engineering, and harmful voice commands. Traditional security methods rely on converting speech to text and then filtering that text,...

Sparse Autoencoders Are Capable LLM Jailbreak Mitigators

Jailbreak attacks remain a persistent threat to large language model safety. We propose Context-Conditioned Delta Steering CC-Delta, an SAE-based defense that identifies jailbreak-relevant sparse features by comparing token-level representations of the same harmful request with and without...