Something I've thought about a lot is that there is having someone with some domain knowledge or reason to care a lot about a particular issue spend a bunch of tokens and cycles on it until something useful comes out the other end. The most obvious ones are the math problems that have been coming out and help push the frontier of various areas of math. Another example is taking all of the public NYC open data ecosystem and crunching it to get some value which I have spent a lot of time and tokens on but not found a great medium to share.
The question is just how to organize these outputs and conclusions in a way that is consistently reproducible and also how to correct errors or remove LLM nonsense where it refuses to take a position on something.
Before it made sense to do this in papers but it feels like we need something like a paper format.. that is fully reproducible ideally and optimized for aggregating knowledge in a better way. I.e. before a person spent months on one of these and there was just more filtering, and the output itself was a clear signal of time spent and effort that no longer exists.
Yes, I’ve thought about this too. The strength of these models is that there is a lot more knowledge encoded in them than the average scientist has in mind at any given time. That means they can explore many more possible combinations of concepts.
If we imagine a set of all human ideas that these models have access to, then the set of possible discoveries would be something like the superset of all possible combinations of those ideas. I think all LLM discoveries are bounded by that space.
Looking at the recent OpenAI math discoveries, that seems to be pretty much what happened. Existing ideas were used as building blocks, the model found a valuable combination, and the result was something new that had real value.
Yeah and I think even what might be the most useful is extracting how Codex got to a particular solution into a skill or into even more customized software so it can be applied in lots more places.
I know people are working on these things I just haven’t seen the right way yet. Like in manufacturing right now people are trying to encode what skilled machinists do into software and scale it up, we need to go further on that for Math/ data analysis etc.