RAG-FiT
Open-source framework that helps train large language models to perform better at retrieval-augmented generation workflows.
RAG-FiT is an open-source library from Intel Labs that helps large language models make better use of external information through fine-tuning specifically for retrieval-augmented generation (RAG). The library helps create the data for training, given a RAG technique, helps easily train models using parameter-efficient finetuning (PEFT), and finally can help users measure the improved performance using various, RAG-specific metrics. The library is modular, workflows are customizable using configuration files.
Explore the project on GitHub (IntelLabs/RAG-FiT) and its accompanying paper, “RAG Foundry: A Framework for Enhancing LLMs for Retrieval Augmented Generation”. The code is released under the Apache 2.0 license and includes practical tutorials, to illustrate end-to-end usage.