alan-turing-institute / autoemulate

emulate simulations easily
MIT License
18 stars 3 forks source link

Experimental: LLM+RAG backbone #164

Open mastoffel opened 9 months ago

mastoffel commented 9 months ago

The idea is to have an option AI=True, which would do the following:

To do this, we can use RAG. For example, we can include a document detailling each model and parameter and details about the output format needed for autoemulate. Structured json output + RAG should make this fairly robust.