Rioghasarig

Does GPT-3 need to understand the world to be useful? If you use it the right way, it still may be possible to accomplish some impressive things.

Yes, it's easy for anyone with a modicum of knowledge about scientific integrity. Wonder why OpenAI doesn't seem to get it.

Yeah, me too. I was really starting to be convinced of GPT-3's potential but this is a major red flag for me. Are they researchers or businessmen?

This is the first time I've heard this. This kind of thing definitely needs more news coverage.

I actually just sent out an API request for a particular application that I believe GPT-3 should be capable.

I'm a university mathematics instructor. I find that a lot of students struggle with doing proofs and I think GPT-3 can help with that. A proof is essentially a sequence of logical inferences. I believe that, when given a proof written in natural language GPT-3 should be capable of detecting the logical implications made within. From there, it would be possible to correct student's mistakes or help them improve their answer.

I don't now about that. I believe if you had it calibrated correctly no one could tell the difference between a single GPT-3 comment and a human comment.

Maybe after a bit of dialogue you'd have a higher chance, but even then I suspect (from playing around in AI Dungeon) that GPT-3 could do very well.

I think pattern matching can be interpreted as a form of reasoning. But it is distinct from logical reasoning. Where you draw implications from assumptions. GPT seems really bad at this kind of thing. It often outputs texts with inconsistencies. And in the GPT-3 paper it performed poorly on tasks like Recognizing Textual Entailment which mainly involves this kind of reasoning.

> The only question it appears to answer is, "given some words, what are other words that are likely to follow them in a typical blog post?" The fact that the words are syntactically correct is unimportant, when the fluent words convey no information relevant to the input.

You say that like that's a bad things. That's literally all it's been trained to do.

Thank you for saying this. This is something so many people miss when trying test the limitations of GPT-2. It just doesn't make sense to test it on strings of text that nobody ever writes.

Why doesn't the fact that it responds with a number represent some level of understanding?

It looks to me you are pointing out one thing that GPT-2 can't understand and declaring "It doesn't understand anything!" while completely ignoring all the things it can be said to understand.

I can't agree with you here. While GPT-2 isn't good at filling in particular details (like the language someone from Boston should speak) it is astonishingly good at recognizing the _kind_ of answer that should be produced. The fact that it usually answers with a language here is a reflection of what it understands. And it behaves similarly for a whole range of different tasks. If you write a sentence that should obviously end with the name of a person it will give you the name of a person. If you write something that should be completed by the name of an event it will give you the name of an event. If you write text that should obviously be finished by the description of a person's appearence it will give you a description of a person's appearence. GPT-2's grasp of the appropriate context of a wide range of situations is nothing short of incredible. Dismissing this as "Eliza on steroids" is just short-sighted.

From what I understand about GPT-2 you shouldn't expect the story to make coherent sense. It can produce individual responses that are surprisingly appropriate but it doesn't actually have an understanding of anything that is happening.