context-segmentation

Segmentación de Contexto para Agentes SLM Esta es la implementación de la solución propuesta en el artículo científico: Evaluating Context Segmentation in Locally Deployable SLMs for Cybersecurity CTF Tasks Configuración Asegúrate de tener Python 3.12.3+ instalado junto con Docker engine Entorno ...

AgentWatcher: A Rule-Based Prompt Injection Monitor

Large language models LLMs and their applications, such as agents, are highly vulnerable to prompt injection attacks. State-of-the-art prompt injection detection methods have the following limitations: 1 their effectiveness degrades significantly as context length increases, and 2 they lack...