Recently I was researching how incorrect explanations of product behavior creep into answers generated by AI answer engines when I came across an interesting finding of value to technical writers.
Researchers at Ericsson were testing retrieval-augmented generation (RAG) against technical documents. One of their discoveries was that keywords appearing closer to the beginning of a sentence were retrieved by AI tools with high accuracy. Keywords appearing later in sentences were harder for AI to retrieve accurately.
Before The Model Answers Anything
The paper, Observations on Building RAG Systems for Technical Documents (PDF download at end of article), was written by Sumit Soman and Sujoy Roychowdhury. The researchers tested technical info from IEEE wireless networking specifications and a battery terminology glossary — reference documents, not how-to step-by-step instructions. They examined factors including chunk length, retrieval method, definitions, and keyword position.
Their keyword-position findings:
Keywords closer to the beginning of a sentence are retrieved with high accuracy.
Keywords appearing later in a sentence were more difficult to retrieve.
