Fine-tuning
Fine-tuning adapts a pretrained model through further training. Supplying current documents at runtime is a different operation.
Fine-tuning continues training a pretrained model with data for a particular task or domain. It adapts model behaviour rather than merely adding a document to each request.
An answer format example
A hypothetical team wants a model to answer service requests in a defined short format. Suitable training examples might teach that behaviour. Reliability must be assessed using separate evaluation examples.
Fine-tuning is not the same as RAG: retrieving current sources is another function. The approaches can be combined. Training replaces neither maintenance of valid work instructions nor access checks. Adapted output still needs evaluation for factual correctness and unwanted behaviour.
Sources
Related terms
- Large language modelA large language model processes language using learned patterns. Fluent text proves neither factual accuracy nor access to current business data.
- Retrieval-augmented generationRAG combines information retrieval with generated answers. Access to sources does not remove the need for quality and permission design.
- LLM evaluationLLM evaluation tests model outputs against defined tasks and criteria. One good example does not establish overall quality.
- Model cardA model card documents an AI model’s purpose, foundations, evaluation conditions and known limitations for a particular version.
- AI readinessAI readiness concerns the foundations of an AI project: objectives, usable data, skills, responsibilities and an operating workflow.
- Context engineeringContext engineering prepares information for individual model calls: relevant instructions, messages, sources and current states.
Articles on this topic
AI & Automation with ASES
AI assistants and automated workflows