Combining dense embeddings with BM25 for advanced local LLM RAG pipeline
This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. RAG systems combine information retrieval with generative models to provide accurate and contextually rich responses.
Andrew Ng has launched a new short course on embedding models, covering their history, architecture, and capabilities. The course, taught by Vectara's Ofer Mendelevitch, explores word, sentence, and cross-encoder models, BERT training, and building dual encoder models for semantic search.
txtai is an open-source embeddings database for various applications such as semantic search, LLM orchestration, language model workflows, and more. It allows users to perform vector search with SQL, create embeddings for text, audio, images, and video, and run pipelines powered by language models for question-answering, transcription, translation, and more.
This article explains Retrieval Augmented Generation (RAG), a method to reduce the risk of hallucinations in Large Language Models (LLMs) by limiting the context in which they generate answers. RAG is demonstrated using txtai, an open-source embeddings database for semantic search, LLM orchestration, and language model workflows.
