Comment by vatsachak

No that's not sufficient but that's not what's happening.

Why do we believe that we cannot train models which could explain the jargon in more human terms when current LLMs can perfectly explain the most complicated codebases?