Re: [xsl] ChatGPT front-end to XML?

["Martynas Jusevičius martynas@xxxxxxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

You are describing RAG here (Retrieval-Augmented Generation).

The most promising approach is hybrid RAG IMO, which combines:
- LLM-generated structured queries (in this case probably XQuery) that
would be able to give exact answers to questions like "What airport
has the longest runway?"
- vector similarity search over embedded documents (for example by
embedding the XML documents using OpenAI embedding models) which would
enable answers to questions like "what are the airports that are
hardest to land in"?

The LLM acting as agent would orchestrate those "tools" and "join"
their answers. You could ask hybrid questions such as "Are airports
with long runways hard to land in"?
My examples might not be the best, I came up with them on the spot :)

I have done quite promising PoC using RDF and SPARQL, not XML.
https://www.ontotext.com/knowledgehub/fundamentals/what-is-graph-rag/

On Thu, Dec 26, 2024 at 12:32b/AM Roger L Costello costello@xxxxxxxxx
<xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:
>
> Hi Folks,
>
> I have an XML document containing data about all the airports in the world.
>
> I want to put a ChatGPT front-end to the XML document, so that users can ask
questions about its data, such as, "What airport has the longest runway?" or
"Please provide a list of civilian airports in France."
>
> Have you put a ChatGPT front-end to an XML document? What lessons did you
learn?
>
> /Roger