I think they show that additional prompting for testing approaches have wide differences in error rate (worst has twice the error rate of the best) but actually no extra prompting is pretty fine and most custom prompts are worse than no prompt.
Regulation might have something to do with it but your example doesn't show that imo.
Multiple US labs were taken to court over copyright infringment for their training data. Mistral did the same data mining, but actually the copyright situation is weaker here so it's harder to build a case against them.
I think the capital situation in EU is just worse. A lot of is tied up in more conservative businesses that are reluctant to bet the house on some shiny new thing.
They are, as are plenty of folks in the community. I expect to see some good talks on this at the AMS Annual Meeting in January, based solely on my own peer network and what colleagues have mentioned they're working on.
This line of thinking is quite confusing to me. If you keep following it then ultimately we must reckon with the reality that we only experience time in one direction. While it might be technically true that "there is only so much you can do with that" I think it might also be a completely useless statement to make.
I think it provides a necessary pointer to the idea that we will need more than prediction to make a viable general AI - that the prediction model is useful, but has fundamental limitations.
I think they show that additional prompting for testing approaches have wide differences in error rate (worst has twice the error rate of the best) but actually no extra prompting is pretty fine and most custom prompts are worse than no prompt.
reply