Large Language Models as a Building Blocks
16Check out my essays: https://aisc.substack.com/
OR book me to talk: https://calendly.com/amirfzpr
OR subscribe to our event calendar: https://lu.ma/aisc-llm-school
OR sign up for our LLM course: https://maven.com/aggregate-intellect/llm-systems
🟢 Search systems can be improved by using language models to understand the meaning of queries and documents, rather than just matching keywords.
🟢 Semantic search can be broken down into two main components: dense retrieval and ranking. Dense retrieval uses embeddings to find similar documents, while ranking uses a language model to score the relevance of each document to the query.
🟢 Retrieval augmented generation (RAG) is a technique that combines search and generation. In RAG, a search system is used to find documents that are relevant to a query, and then a language model is used to generate text that summarizes or answers the query based on those documents.
🟢 RAG can be improved by using query rewriting to reformulate the user's query into a more searchable form.
🟢 Large language models (LLMs) are still under development, and it is important to be aware of their limitations. For example, LLMs can sometimes generate text that is incorrect or misleading.