Context stuffing
Context stuffing is pasting large amounts of material into a prompt in the hope that the model will find the relevant part itself.
It is usually counterproductive. More context costs more, slows responses, and dilutes attention across irrelevant material, and anything placed mid-stack is the least likely to be recalled. Selecting the right 500 words beats supplying the wrong 50,000.
Related terms
- Lost in the middle
- Lost in the middle is the observed effect that models recall information placed at the start or end of a long context more reliably than information in the middle.
- RAG (retrieval-augmented generation)
- RAG is a technique that retrieves relevant documents from an external store and places them in the prompt, so the model answers from that material rather than from memory.
More on prompt structure
See the full glossary, read the guides, or put it into practice in the prompt builder.