Hacker News new | past | comments | ask | show | jobs | submit login

Show the training set, and PROVE that the tasks and answers aren't in there. I don't understand why this is not a default first step for proving that this is creating new knowledge.





Well that's harder than maybe solving well-known open problems (whose soln's are presumably not in training set lol) but it seems that their examples are not clearly breaking sota, especially on matmul

Are you claiming that for the open problems they give record-breaking solutions for, there were just answers on the web waiting to be found?

No, I'm saying they have a massive database of solutions (the training set) and don't even bother proving that their solution isn't in there. I'm not claiming something, they are failing to provide some necessary information here

It's Google. Assume the training set contains, as a subset, the entirety of all public digitized information. How would you like to them to share it?

If they wanna do research where they claim they did something novel, without showing that they didn't just "look it up" in their massive training set, then yes, they should share what is and what isn't contained within.

As a tar, please.

How can you actually verify it, even if they provide something?

That's my point; you can't. They have no idea if their model came up with any of this or not.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: