Production Guardrails for Compressed LLMs: Confidence and Abstention
Learn how production guardrails for compressed LLMs use confidence scores and abstention to balance safety and speed. Explore Defensive M2S, efficiency techniques, and implementation strategies.
Retrofitting Transformers with Guardrails: Safety Layers for Enterprise LLMs
Learn how retrofitting transformers with guardrails creates essential safety layers for enterprise LLMs. Explore defense strategies against prompt injections, compliance with EU AI Act, and tools like OneShield.
Safety-Aware Prompting: How to Prevent Sensitive Data Leaks in GenAI
Learn how to use safety-aware prompting to prevent data leaks and prompt injections in Generative AI. Practical habits and technical strategies for secure LLM use.
