Software evolution/changes over time in:

  • complexity
  • scale
  • number of people involved
  • productivity
  • costs

M3: meta meta model M2: meta model M1: model

Q. ensure easy and correct co-evolution between code and models. How to evaluate the co-evolution?

GumTree: algorithm to detect tree differences

4 contributions


External Q. Generalize the approach to languages other than Java (which is the most used for software modelization)

External Q. Direct applications in the industry, use cases, examples

External Q. Methodology aspects in experiments. Sample size, inclusion criteria. Taking in account fatigue, learning curve into experiments?

External Q. How did you get to the final set of prompts used to evaluate LLM output?

External Q. Automation is always good, but there are scenarios in which a human intervention is necessary? why?

External Q. Precisation about terminology