Translator Feedback Loop for Joomla (Inspired by RLHF)

This project introduces a native Translator Feedback Loop for Joomla (Inspired by Reinforcement Learning from Human Feedback "RLHF") to improve machine translations over time. This project will work by capturing human corrections and injecting them into future LLM prompts via a decoupled RAG architecture, the system aims to ensure consistent, community-standard terminology for all the supported languags. The Final deliverable is a extension package including a specialized translator dashboard, background task scheduler and AI framework integration.