察元AI教程库 · 仓库 · 提示词
An advanced RAG pipeline to query and evaluate the UMRI Informatics Thesis Guidebook. Powered by LlamaParse, Qdrant, NVIDIA NIM, and cross-encoder re-ranking. Features Prompt Engineering (CoT vs Non-C…
作者:anisadwisww · 语言:en
An advanced RAG pipeline to query and evaluate the UMRI Informatics Thesis Guidebook. Powered by LlamaParse, Qdrant, NVIDIA NIM, and cross-encoder re-ranking. Features Prompt Engineering (CoT vs Non-CoT) and rigorous pipeline evaluation using RAGAS metrics (Faithfulness, Answer Relevancy, Context Precision, Context Recall).