EduRAG
An intelligent question-answering system for the IT domain, built on the RAG (Retrieval-Augmented Generation) architecture and combining an internal knowledge base to provide professional answers. It comprises a core RAG module and a FAQ cache acceleration module.
Timeline (last 90 days)
Heima Programmer launched a project course on the EduRAG intelligent Q&A system, with instructor Xing Bu (currently a core member of an AI large model development team, formerly at Didi, ByteDance, and TAL), and the course is released based on Bilibili's official videos.
The EduRAG project adopts a dual-module architecture: the RAG core module (including the three processes of retrieval, augmentation, and generation) and the FAQ module (which reduces response latency and Token consumption by caching high-frequency question-answer pairs).
The EduRAG project course was originally planned for 10 days and was compressed to 8 days: days 1-2 background and tools, day 3 Milvus and LangChain, days 4-5 the FAQ module, days 6-8 the RAG core, and finally project integration and Docker deployment, followed by the EDU Agent project.
The project involves a model fine-tuning step that performs binary classification on user input, used to determine whether it is a professional question or an irrelevant question, allowing only professional questions into the RAG system to control Token consumption, an approach known in industry as query routing or intent classification.
The project loads the original model files via the Transformers framework to run Embedding on local compute, no longer relying on the platform's one-click conversion.
All Facts (5)
The EduRAG project adopts a dual-module architecture: the RAG core module (including the three processes of retrieval, augmentation, and generation) and the FAQ module (which reduces response latency and Token consumption by caching high-frequency question-answer pairs).
50%UnverifiedThe EduRAG project course was originally planned for 10 days and was compressed to 8 days: days 1-2 background and tools, day 3 Milvus and LangChain, days 4-5 the FAQ module, days 6-8 the RAG core, and finally project integration and Docker deployment, followed by the EDU Agent project.
50%UnverifiedThe project involves a model fine-tuning step that performs binary classification on user input, used to determine whether it is a professional question or an irrelevant question, allowing only professional questions into the RAG system to control Token consumption, an approach known in industry as query routing or intent classification.
50%UnverifiedThe project loads the original model files via the Transformers framework to run Embedding on local compute, no longer relying on the platform's one-click conversion.
50%UnverifiedHeima Programmer launched a project course on the EduRAG intelligent Q&A system, with instructor Xing Bu (currently a core member of an AI large model development team, formerly at Didi, ByteDance, and TAL), and the course is released based on Bilibili's official videos.
50%