RAG (Retrieval-Augmented Generation)
定義
Retrieval-Augmented Generation (RAG) is a technique that enhances LLM outputs by retrieving relevant documents from a knowledge base and including them as context in the prompt. Instead of relying solely on the model's training data, RAG grounds responses in your organization's specific documents, policies, and data — reducing hallucination and increasing accuracy.
How RAG Works in JieGou
Upload documents (PDF, DOCX, Markdown, HTML, or URLs) to a knowledge base. JieGou chunks documents into passages, generates embeddings, and stores them. When a recipe runs, relevant chunks are retrieved by semantic similarity and injected into the prompt as context. The LLM sees your data alongside the task instructions.
Auto-Context
Knowledge bases can be scoped to departments, recipes, or workflows. Auto-context automatically includes relevant knowledge bases based on the scope — a Sales recipe automatically gets context from the Sales knowledge base without manual configuration.
関連用語
AIレシピ
AIレシピとは何か、JieGouでどのように機能するかをご紹介します。レシピは構造化された入出力を持つ、再利用可能な単一操作のAIビルディングブロックです。
Prompt Template
A prompt template is a reusable, parameterized set of instructions for an LLM that accepts variable inputs and produces structured outputs.
Large Language Model (LLM)
A large language model (LLM) is an AI system trained on text data that can understand and generate human language, powering tasks like writing, analysis, and reasoning.