A deep dive into superior indexing, pre-retrieval, retrieval, and post-retrieval strategies to reinforce RAG efficiency
10 hours in the past
Have you ever requested a generative AI app, like ChatGPT, a query and located the reply incomplete, outdated, or simply plain incorrect? What if there was a option to repair this and make AI extra correct? There’s! It’s referred to as Retrieval Augmented Era or simply RAG. A novel idea launched by Lewis et al of their seminal paper Retrieval-Augmented Era for Data-Intensive NLP Duties , RAG has swiftly emerged as a cornerstone, enhancing reliability and trustworthiness within the outputs from Massive Language Fashions (LLMs). LLMs have been proven to retailer factual data of their parameters, additionally known as parametric reminiscence and this information is rooted within the information the LLM has been educated on. RAG enhances the data of the LLMs by giving them entry to an exterior info retailer, or a data base. This information base can also be known as non-parametric reminiscence (as a result of it isn’t saved in mannequin parameters). In 2024, RAG is among the most generally used strategies in generative AI functions.
60% of LLM functions make the most of some type of RAG