mehalter
commented
4 years ago I added a problem statement that semantically sound model fusion becomes more difficult as models scale and you are understanding source code manually, and then followed it up in the results that semantic models allows you to do semantic model fusion, and it handles fusing the math toegehter automatically
I think the poster needs to talk more about the context in terms of CSE problems.
The CSE/PP audience will appreciate the difficulty of model fusion, model selection, multi scale solvers, model order reduction , and optimizing compilers that exploit model structure. So framing our approach in terms of solutions to those problems would be good.
If we had some result about generating parallel code from the PetriNet description or doing some other optimization in the Structured -> Code transformation, that would be really compelling to this audience.