"Databricks has an answer to that problem, Instructed Retriever, which breaks down requests into specific search terms and filter instructions when retrieving documents to augment the generative prompt. That means, for example, that a request for product information with an instruction to "focus on reviews from the last year" can explicitly retrieve only reviews for which the metadata indicates they are less than a year old."
"Everything about retrieval-augmented generation (RAG)'s architecture was supposed to be simple. It was the shortcut to enterprise adoption of generative AI: retrieve documents that may be relevant to the prompt using similarity search, pass them to a language model along with the rest of the prompt, and let the model do the rest. But as enterprises push AI systems closer to production, that architecture is starting to break down."
Databricks introduced Instructed Retriever, an architecture that augments RAG by translating user instructions into explicit search terms and metadata filters. The approach combines deterministic database queries with similarity search to return more relevant and constrained document sets. Instructed Retriever can restrict results to reviews whose metadata indicates they are less than a year old when a prompt specifies “focus on reviews from the last year.” Traditional RAG treats such instructions as part of the prompt and relies on the model to reconcile retrieved documents, creating trade-offs among latency, accuracy, and control as enterprises move systems toward production. Analysts warn CIOs that the method may expose gaps in data, governance, and budget.
Read at InfoWorld
Unable to calculate read time
Collection
[
|
...
]