In Achieving High-Level Software Component Summarization via Hierarchical Chain-of-Thought Prompting and Static Code Analysis (ICoDSE'23) we presented an approach for automated software component summarization, combining static analysis with the emerging capabilities of a Large Language Model (LLM). In short, we use LLM to generate summaries of individual Java methods from their source code and then gather the summaries of all methods under a class to then generate a summary of that class. Analogously, we gather the summaries of all classes under a package to then generate a summary of that package.
(In the figure, operation is synonymous with method, module with class, and component with package.)
A showcase for this technique is shown here. Everything in that page is generated from pure source code, i.e., discarding any sorts of comments. An evaluation for this technique is currently being planned.