Adaptive_Greedy_Local_Search

Адаптивный жадный локальный поиск AGLS Семантически сохраняющий угон промптов: атака черного ящика на автоматическую оптимизацию промптов ICME 2026 Аннотация: Крупные языковые модели LLM все чаще оснащаются модулями автоматической оптимизации промптов, которые переписывают ввод пользователя и явн...

sk-cve-2026-26030-lab

CVE-2026-26030 — RCE через фильтр eval в Semantic Kernel лаборатория Автономная лаборатория, воспроизводящая CVE-2026-26030 : внедряемое через подсказку удаленное выполнение кода через фильтр поиска в векторном хранилище в оперативной памяти Microsoft Semantic Kernel Python, Только этическая...

FlowName

FlowName Name quality shows the ratio of names exclusively preferred by an automated, uncalibrated Jev evaluation.Animated chart → · Exact counts, method and limitations → Try FlowName in your browser → — enter your JavaScript, API key and model; no installation required. CLI: npx --yes...

SwordStamp

SwordStamp SwordStamp adds order-robust detection and semantic-span units to embedding-based semantic watermarks. This artifact contains the exact paper configuration, generation and detection code, the unchanged attack suite, quality evaluation, and deterministic result extraction and plotting...

ablation

Ablation — это фреймворк для обратной разработки, предоставляющий ровно те же базовые возможности дизассемблирования, декомпиляции и анализа бинарных файлов, что и инструменты отраслевого стандарта, такие как Ghidra, IDA Pro и Binary Ninja. В сочетании с Claude Code или OpenAI Codex он превращает...

DeepGuard

DeepGuard 📖 프로젝트 개요 DeepGuard는 다중 계층 의미 집계multi-layer semantic aggregation 기법을 통해 대규모 언어 모델의 안전한 코드 생성 능력을 향상시키는 혁신적인 보안 코드 생성 접근 방식입니다. 이 방법은 코드의 보안 취약점을 효과적으로 식별하고 완화하여 개발자에게 더 안전한 코드 생성 솔루션을 제공합니다. 🔑 핵심 기술 특징 다중 계층 의미 집계Multi-Layer Semantic Aggregation : 여러 Transformer 계층의 은닉 상태를 집계하여 풍부한 의미 정...

React2Shell-CVE-2025-55182-Detector

Предварительные требования Доступ к корневому каталогу проекта, содержащему package.json Одна из следующих сред выполнения, в зависимости от используемого скрипта: Python Javascript PowerShell Bash с установленным jq Использование Все скрипты должны выполняться из корневого каталога целевого...

[SECURITY] Fedora 45 Update: sparse-0.6.4-4.gce1a6720f69e.fc45.9

Sparse is a semantic parser of source files: it's neither a compiler although it could be used as a front-end for one nor is it a preprocessor although it contains as a part of it a preprocessing phase. It is meant to be a small - and simple - library. Scanty and meager, and partly because of tha...

CVE-2026-26030-Microsoft-Semantic-Kernel-1.39.4-RCE

Microsoft Semantic Kernel BPassed to VectorStore query/filter B -- CInMemoryVectorStore evaluates filter expression unsafely C -- DUnsafe attribute resolution via class / base / subclasses D -- EAccess to init.globals containing os and builtins E -- FArbitrary code execution e.g. os.system F --...

CVE-2025-66698

CVE-2025-66698 Целевое ПО: Veda Semantic Machines v5.4.8 вероятно, все предыдущие версии. Тип уязвимости: Обход аутентификации. Идентификатор CVE: CVE-2025-66698. Краткое описание: Система некорректно проверяет параметр ticket, считая пустую строку действительной сессией. Ссылка на проект: Veda...

A Resource-Aware Behavior Reconstruction and Hierarchical Semantic Learning Framework for Host Intrusion Detection

System calls syscalls record key interactions between running programs and the operating system kernel, providing fine-grained and minimally intrusive data for host-based intrusion detection systems HIDS deployed in cloud and other modern computing environments. However, existing methods often...

SUSE CVE-2026-89032

BerriAI LiteLLM before 1.101.0-rc.1 contains a tenant isolation bypass vulnerability in the semantic cache layer that allows authenticated users to read other tenants' cached responses by exploiting a metadata key mismatch between getsemanticcachetenantscope and getmetadatavariablename. Attackers...

Incorrect Authorization

Overview litellm is a Library to easily interface with LLM API providers Affected versions of this package are vulnerable to Incorrect Authorization via a tenant isolation bypass vulnerability in the semantic cache layer that allows authenticated users to read other tenants' cached responses by...

CVE-2026-89032

BerriAI LiteLLM before 1.101.0-rc.1 contains a tenant isolation bypass vulnerability in the semantic cache layer that allows authenticated users to read other tenants' cached responses by exploiting a metadata key mismatch between getsemanticcachetenantscope and getmetadatavariablename. Attackers...

CVE-2026-89032

A flaw was found in LiteLLM. Due to a metadata key mismatch in the semantic cache component, an authenticated remote attacker can bypass tenant isolation by submitting semantically similar queries. This issue leads to information disclosure, allowing unauthorized access to cached responses...

CVE-2026-89032 BerriAI LiteLLM < 1.101.0-rc.1 Tenant Isolation Bypass via Semantic Cache Layer

BerriAI LiteLLM before 1.101.0-rc.1 contains a tenant isolation bypass vulnerability in the semantic cache layer that allows authenticated users to read other tenants' cached responses by exploiting a metadata key mismatch between getsemanticcachetenantscope and getmetadatavariablename. Attackers...

EUVD-2026-87147

BerriAI LiteLLM before 1.101.0-rc.1 contains a tenant isolation bypass vulnerability in the semantic cache layer that allows authenticated users to read other tenants' cached responses by exploiting a metadata key mismatch between getsemanticcachetenantscope and getmetadatavariablename. Attackers...

CVE-2026-89032

BerriAI LiteLLM versions prior to 1.101.0-rc.1 are vulnerable to a tenant isolation bypass within the semantic cache layer . The flaw is caused by a metadata key mismatch between the _get_semantic_cache_tenant_scope() and _get_metadata_variable_name() functions. An authenticated remote attacker w...

CVE-2026-89032 BerriAI LiteLLM < 1.101.0-rc.1 Tenant Isolation Bypass via Semantic Cache Layer

BerriAI LiteLLM before 1.101.0-rc.1 contains a tenant isolation bypass vulnerability in the semantic cache layer that allows authenticated users to read other tenants' cached responses by exploiting a metadata key mismatch between getsemanticcachetenantscope and getmetadatavariablename. Attackers...

CVE-2026-89032 BerriAI LiteLLM < 1.101.0-rc.1 Tenant Isolation Bypass via Semantic Cache Layer

BerriAI LiteLLM before 1.101.0-rc.1 contains a tenant isolation bypass vulnerability in the semantic cache layer that allows authenticated users to read other tenants' cached responses by exploiting a metadata key mismatch between getsemanticcachetenantscope and getmetadatavariablename. Attackers...