| ▲ | aucisson_masque 3 hours ago | |
> Large language models want data about coding problems and how to solve them. Stack Overflow has a big digital warehouse full of that, but it’s increasingly aging, as queries move into private chat windows with LLM models. What are the llm going to feed on when coding languages change and there isn't anymore stackoverflow or these kind of forum ? Surely it can read the documentation but it's not enough, you need data from real humans figuring stuffs out. | ||
| ▲ | xtracto 2 hours ago | parent [-] | |
Within a couple of years, LLMs or equivalent AI engines will have the ability to generate factual knowledge themselves using their "outdated" knowledge and their reasoning mechanisms. For example, one of the classical requests from S.O. questions and GitHub issues is a "minimal reproducible example " of the question/problem. So a sufficiently advanced AI will be able to write that, run it, see the issue , go to the library/related-system code or documentation (for closed source) and derive a solution | ||