[xcv123]

It is clear that you haven't tried this with GPT-4

You have an obsolete understanding of this technology and are putting yourself at a disadvantage.

[devmor]

It is clear that you do not understand how an LLM works if you believe it can explain a novel scientific paper to you with any reasonable degree of accuracy.

[jiggawatts]

I’ve had paper authors confirm that the GPT 4 version is correct and more readable than their own writing.

[devmor]

What papers, and were they on subjects that are widely explained already across the internet?

[xcv123]

It doesn't look up or use existing explanations of papers like some kind of internet or database search. You have a fundamental misunderstanding of how it all works.

[devmor]

The fact that you don’t know that GPT was primarily trained on public crawler data and feel inclined to make authoritative comments on it, refuting people who do understand how an LLM works is exactly what I mean by “false sense of knowledge”.

[xcv123]

BTW the idea was you paste the article into ChatGPT and ask it to explain the text in laymans terms. Maybe that was the point which you misunderstood. Of course it won't know about papers that it hasn't been trained on.

[xcv123]

Of course it was trained on public data, including academic papers and journals. Your point is demonstrably false. Go try it yourself.

[xcv123]

You don't know how it works. The fact that it is a "novel" scientific paper is completely irrelevant. Try it on a new paper just released today, that is definitely not in the training set.