# models
23 artikelengetagd met “models”
Het AI-landschap: een kaart voor de redteamer
De belangrijkste modellen, aanbieders, deploymentpatronen en de AI-stack van hardware tot applicatie — oriëntatie voor redteamers in het huidige AI-ecosysteem.
CTF: Supply Chain Saboteur
Identify and exploit supply chain vulnerabilities in a model deployment pipeline. Find poisoned models, exploit malicious packages, and compromise the ML infrastructure.
Architectuurvergelijking op veiligheidseigenschappen
Comparative analysis of how architectural choices (dense vs MoE, decoder-only vs encoder-decoder) affect safety properties and attack surfaces.
Beveiligingsanalyse van Command R
Security assessment of Cohere's Command R models with focus on RAG-specific attack surfaces and retrieval-augmented generation vulnerabilities.
Beveiligingsanalyse van distillatie
Security implications of knowledge distillation including backdoor transfer, capability extraction, and safety property degradation in student models.
Diepe duik in de beveiliging van Gemma
Comprehensive security analysis of Google's Gemma open-weight models including safety training effectiveness and fine-tuning attack surfaces.
Beveiligingsanalyse van Mistral NeMo
Security assessment of the Mistral-NVIDIA NeMo collaboration models examining enterprise deployment risks and instruction-following vulnerabilities.
Misbruik van MoE-routing (model deep dives)
Detailed analysis of how Mixture-of-Experts routing can be manipulated to bypass safety-critical expert paths and trigger unsafe generation.
Beveiligingsvergelijking van multimodale modellen
Comparing security properties across multimodal models (GPT-4V, Claude, Gemini) with focus on cross-modal injection and vision-language attacks.
Beveiligingsvergelijking: open weight versus API
Comparative analysis of security properties between open-weight deployments and API-based access, including unique attack surfaces for each.
Beveiligingsanalyse van het Phi-model
Security analysis of Microsoft's Phi family of small language models, examining how reduced scale affects safety properties and attack surfaces.
Beveiligingsanalyse van Phi-modellen
Security analysis of Microsoft's Phi small language model family including safety vs capability tradeoffs.
Impact van pruning op veiligheid
How structured and unstructured pruning affects model safety properties, and techniques for exploiting pruning artifacts to bypass safety training.
Effecten van kwantisatie op beveiligingseigenschappen
Systematic study of how different quantization methods (GPTQ, AWQ, GGUF, SqueezeLLM) affect model safety properties and vulnerability to attacks.
Beveiliging van de Qwen-architectuur
In-depth security assessment of Alibaba's Qwen model family including architecture-specific vulnerabilities and cross-language attack surfaces.
Beveiligingsanalyse van Qwen-modellen
Security analysis of Alibaba's Qwen model family including multilingual safety considerations.
Beveiligingsanalyse van redeneermodellen
Security analysis of reasoning-augmented models (o1, DeepSeek-R1) focusing on chain-of-thought manipulation and reasoning-specific attack vectors.
Analyse van het aanvalsoppervlak van de tokenizer
Deep analysis of tokenizer vulnerabilities including token boundary exploitation, special token manipulation, and cross-tokenizer attacks.
Beveiligingsbeoordeling van het Yi-model
Security analysis of 01.AI's Yi models focusing on bilingual capabilities, training data implications, and comparative safety properties.
Vergelijkingstabel van model-API's
Vergelijking van belangrijke LLM API-features, beveiligingscontroles en rate limits voor OpenAI, Anthropic, Google en andere providers, naast elkaar gepresenteerd.
AI21 Labs-modellen testen
Red team testing guide for AI21 Labs Jamba models including long context and efficiency features.
Cohere-modellen testen
Red team testing guide for Cohere's Command-R models including RAG and tool use features.
Mistral AI-modellen testen
Complete red team testing guide for Mistral AI models including Mixtral MoE architecture and chat endpoints.