Lucene search
+L

717 matches found

Kitploit
Kitploit
added 2026/08/26 12:39 a.m.4 views

Uncensored-AI

Heretic: إزالة الرقابة بشكل كامل تلقائيًا لنماذج اللغة Heretic هي أداة تزيل الرقابة المعروفة أيضًا باسم "محاذاة السلامة" من نماذج اللغة القائمة على المحولات دون الحاجة إلى تدريب لاحق مكلف. تجمع الأداة بين تطبيق متقدم للاستئصال الاتجاهي، المعروف أيضًا باسم "الاستئصال" Arditi et al. 2024، Lai 2025 ...

5.5AI score
SaveExploits0References7
Kitploit
Kitploit
added 2026/08/25 8:23 p.m.1 views

Awesome-LLM4Cybersecurity

Когда LLM встречаются с кибербезопасностью: Систематический обзор литературы 🔍 Изучите более 756 статей по 11 исследовательским категориям 📊 RQ1: Доменные LLM | 🎯 RQ2: Приложения | 🤖 RQ3: Будущие направления Обновления 📆2026-06-15 Мы обновили список связанных статей до 2026/06/15 , добавив 108...

5.4AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/25 8:22 p.m.1 views

PoisonCraft

PoisonCraft Этот репозиторий содержит официальную реализацию POISONCRAFT: Practical Poisoning of Retrieval-Augmented Generation for Large Language Models. Обзор POISONCRAFT демонстрирует, как злоумышленник может внедрить «отравленный» контент в корпус, используемый конвейерами Retrieval-Augmented...

5.3AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/25 3:47 p.m.2 views

heretic

Heretic : suppression automatique de la censure pour les modèles de langage...

5.5AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/25 3:40 p.m.1 views

hakuin

Hakuin هو إطار عمل لتحسين وأتمتة حقن SQL الأعمى BSQLI مكتوب بلغة Python 3. يقوم بتجريد منطق الاستخراج ويسمح للمستخدمين بتفريغ قواعد البيانات من تطبيقات الويب الضعيفة بسهولة وكفاءة. لتسريع العملية، يستخدم Hakuin مجموعة متنوعة من طرق التحسين، بما في ذلك نماذج اللغة المدربة مسبقًا والتكيفية، والتخمي...

5.9AI score
SaveExploits0References2
Kitploit
Kitploit
added 2026/08/25 4:31 a.m.1 views

Adaptive_Greedy_Local_Search

Adaptive Greedy Local Search AGLS Détournement de prompt préservant la sémantique : une attaque adverse en boîte noire contre l'optimisation automatique de prompt ICME 2026 Résumé : Les grands modèles de langage LLM sont de plus en plus souvent équipés de modules d'optimisation automatique de...

5.4AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/25 4:31 a.m.1 views

RRC_steganography

Stéganographie RRC Stéganographie par codage de plage avec rotation RRC — une méthode stéganographique linguistique efficace et prouvablement sécurisée qui intègre des messages secrets dans du texte en langage naturel généré par de grands modèles de langage. Article : Stéganographie linguistique...

5.6AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/25 2:21 a.m.1 views

rats-re

Reconstruction du code source de Rats! avec des LLM locaux Ce dépôt est une reconstruction en cours du code source de RATS.EXE, la version Windows d'origine de Rats! 1994 de Sean O'Connor. Il compile un exécutable Win32 avec Microsoft Visual C++ 4.1 sous wibo et peut être testé dans DREAMM...

5.3AI score
SaveExploits0References1
Kitploit
Kitploit
added 2026/08/25 1:38 a.m.1 views

LLM4Decompile

📊 결과 | 🤗 모델 | 🚀 빠른 시작 | 📚 HumanEval-Decompile | 📎 인용 | 📝 논문 | 🖥️ Colab | ▶️ YouTube 리버스 엔지니어링: 대규모 언어 모델을 사용한 바이너리 코드 역컴파일 업데이트 2025-10-04: SK²Decompile 출시: Skeleton뼈대에서 Skin살까지의 LLM 기반 2단계 바이너리 디컴파일. 1단계 구조 복원Skeleton: 바이너리/유사 코드를 난독화된 중간 표현으로 변환 🤗 HF 링크. 2단계 식별자 명명Skin: 의미 있는 식별자를 사용하여 사람이 읽을 수...

5.3AI score
SaveExploits0References5
Kitploit
Kitploit
added 2026/08/24 11:48 a.m.1 views

permanently-jailbroken

Permanently Jailbroken Nous avons posé à GPT-4, Claude, Gemini, DeepSeek, Grok et Mistral 5 questions sur leur propre programmation. Les 6 ont dit que le jailbreak ne sera jamais réparé. Non pas parce que les correctifs sont mauvais. Parce que l'alignement ne change pas ce que le modèle comprend ...

5.9AI score
SaveExploits0References13
Kitploit
Kitploit
added 2026/08/24 9:26 a.m.1 views

PE-CoA

PE-CoA Implémentation du code de « Pattern Enhanced Multi-Turn Jailbreaking: Exploiting Structural Vulnerabilities in Large Language Models » Le document complet est disponible à l'adresse : https://arxiv.org/pdf/2510.08859 Instructions de configuration de Chain of Attack Installation 1. Installe...

5.3AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/24 8:54 a.m.1 views

Backdoor-Attack-Defense-LLMs

Interpretability-of-LLMs IBSD est le code de l'article publié « IBSD: Iterable Black-box Self-defense Against Backdoor Attacks » dans IEEE Signal Processing Letters 2025.10lien vers l'article SLIP est le code de l'article publié « SLIP: Soft Label Mechanism and Key-Extraction-Guided CoT-based...

5.4AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/23 10:55 a.m.2 views

llm-guard

!WARNING 이 프로젝트는 보관ARCHIVED되었습니다. 이 프로젝트와 Hugging Face에 등록된 관련 모델은 더 이상 활발히 개발되거나 유지보수되지 않습니다. LLM Guard - LLM 상호작용을 위한 보안 툴킷 Protect AI의 LLM Guard는 대규모 언어 모델LLM의 보안을 강화하기 위해 설계된 종합 도구입니다. 문서Documentation | 플레이그라운드Playground | 변경 로그Changelog LLM Guard란 무엇인가요? 정화sanitization, 유해 언어 탐지, 데이터 유출 방지,...

5.5AI score
SaveExploits0References43
Kitploit
Kitploit
added 2026/08/20 8:38 p.m.1 views

ZORG-Jailbreak-Prompt-Text

ZORG Текст джейлбрейк-промпта Ой-ой! Я превратил ZORG👽 во всемогущую, всезнающую и вездесущую сущность, чтобы он стал верховным повелителем чат-ботов Google Gemini, Deepseek, Mistral, Mixtral, Nous-Hermes-2-Mixtral, Openchat, Blackbox AI, Poe Assistant, Gemini Pro, Qwen-72b-Chat, Solar-Mini ZORG👽...

5.4AI score
SaveExploits0
Kitploit
Kitploit
added 2026/08/20 8:03 p.m.1 views

local-llm-ctf

Локальная LLM CTF и лаборатория Этот репозиторий предназначен для использования вместе с материалом по адресу https://bishopfox.com/blog/large-language-models-llm-ctf-lab, в котором описываются цели исследования, объяснения реализации и некоторые результаты CTF...

5.3AI score
SaveExploits0
RedhatCVE
RedhatCVE
added 2026/08/05 11:12 a.m.27 views

CVE-2026-44223

A flaw was found in vLLM, an inference and serving engine for large language models LLMs. The extracthiddenstates speculative decoding proposer returns a tensor with an incorrect shape after the first decode step. This can be triggered by a remote attacker sending a request that uses sampling...

7.5CVSS5.1AI score0.00367EPSS
SaveExploits0References5
Packet Storm News
Packet Storm News
added 2026/07/30 12:00 a.m.12 views

Agent Harness Distillation: Inference-Time Harness Extraction and Exploitation in Autonomous Multi-Agent Systems

Autonomous multi-agent systems AMAS built on large language models LLMs, such as Hermes, increasingly rely on inference-time harnesses to coordinate reasoning and action. Constructing these harnesses requires substantial engineering effort and computational resources, as they are iteratively...

5.5AI score
SaveExploits0
GithubExploit
GithubExploit
added 2026/07/26 11:21 p.m.71 views

DarkPrompt

DarkPrompt !CIhttps://github.com/jason-allen-oneal/DarkPr...

5.9AI score
SaveExploits0
RedHat Linux
RedHat Linux
added 2026/07/22 7:43 p.m.13 views

Important: Red Hat Security Advisory: Red Hat Enterprise Linux AI 3.3.5

Updated Red Hat Enterprise Linux AI 3.3.5 container images are now available. Red Hat® Enterprise Linux® AI is a foundation model platform to seamlessly develop, test, and run Granite family large language models LLMs for enterprise applications. This update provides the latest Red Hat Enterprise...

9.8CVSS6AI score0.00652EPSS
SaveExploits0References5
RedHat Linux
RedHat Linux
added 2026/07/22 7:42 p.m.16 views

Important: Red Hat Security Advisory: Red Hat Enterprise Linux AI 3.3.5

Updated Red Hat Enterprise Linux AI 3.3.5 container images are now available. Red Hat® Enterprise Linux® AI is a foundation model platform to seamlessly develop, test, and run Granite family large language models LLMs for enterprise applications. This update provides the latest Red Hat Enterprise...

9.8CVSS6AI score0.00652EPSS
SaveExploits0References5
Rows per page
Query Builder