// MODEL OPTIMIZATION AND PROMPT SYNTAX TERM

Retrieval Augmented

Describes AI models or systems that enhance their ability to generate responses by first looking up relevant information from an external source.

TECHNICAL DEFINITION

Retrieval Augmented refers to a class of AI architectures or techniques, prominently Retrieval Augmented Generation (RAG), where a model's generative capabilities are enhanced by integrating an explicit information retrieval component that fetches relevant external knowledge to inform and ground the generation process.

BACKGROUND

Prompt engineering is the process of structuring natural language inputs to produce specified outputs from a generative artificial intelligence (GenAI) model. Context engineering is the related area of software engineering that focuses on the management of non-prompt contexts supplied to the GenAI model, such as metadata, API tools, and tokens.

READ MORE ON WIKIPEDIA

SYNONYMS & ALIASES

  • Knowledge-Enhanced
  • Externally Grounded
  • Information-Informed Generation

USAGE NOTE

Retrieval augmented systems are becoming standard for applications requiring factual accuracy and up-to-date information.

DEVELOPERS

Organizations developing technology related to Retrieval Augmented.

  • LangChain

    A framework designed to simplify the creation of applications using large language models, providing tools for retrieval-augmented generation (RAG) to connect LLMs with external data sources.

  • LlamaIndex

    A data framework for LLM applications that provides tools to ingest, structure, and access private or domain-specific data, making it central to building RAG systems.

  • Hugging Face

    An AI community and platform that provides tools, models, and datasets, including resources and implementations for various RAG architectures and models.

  • Pinecone

    A vector database for building high-performance vector search applications, which serves as a critical component for storing and retrieving embeddings in RAG systems.

  • Weaviate

    An open-source vector database that allows for efficient storage and retrieval of data objects by vector embeddings, foundational for RAG implementations.

  • Google AI

    Conducts extensive research in AI and develops LLMs, often incorporating retrieval mechanisms to ground models with up-to-date or domain-specific information.

  • Meta AI

    Develops and open-sources various AI models and research, including contributions to retrieval-augmented models and techniques for improving LLM factual consistency.

  • Cohere

    Provides enterprise-grade large language models and a platform that supports grounded generation, which is a key aspect of retrieval-augmented approaches.

  • Databricks

    Offers a data intelligence platform that enables enterprises to build and deploy AI applications, including leveraging RAG for custom LLM solutions.

RELATED TERMS IN PROMPTING & LOGIC