Loading…
Loading content
Loading…
Loading contentAn architecture-ready interface for tuning the depth of an answer to its audience — an educational mode for newcomers, a research mode for depth, an expert mode for concision. The modes are defined; a language-model layer would realise them over the same grounded facts.
The interface is defined and modelled in the graph, but no language model is wired in. When one is, it will phrase the grounded facts the retrieval layer supplies — never add to them. Nothing is fabricated.
Facts on this topic will be cited from these primary and reference sources.
Mission data, planetary science, space telescopes, and public-domain imagery.
Most NASA-produced imagery is in the public domain; individual items are checked for usage terms before publication.