News for hackers. Stuff that matters.
Re:Microsoft and OpenAI Working on ChatGPT-Powered Bing in C... by ailef on Wednesday January 04, 2023 @07:31AM (#34244782)
In cases where you ask GPT-3 questions that have a clear correct answer, I think you can use the probability to judge how correct the answer is. For example, when asking "How tall is Mount Everest?" I would want the completion "Mount Everest is ____ meters above sea level." to have a very high probability for the ____ tokens.

This is because I'm operating under the assumption that sequence of words that appear often in the training set are more likely to represent something correct (otherwise you might as well train on random words). This only holds if the training set is big enough that you can compute this estimation correctly (e.g. if the training set is small a very rare phrase may appear very often).

Maybe confidence was the wrong word but for this kind of questions I would trust a high-probability answer way more than the other way. For questions belonging to very specific subjects where training material has not been enough, this is different because as I said the model might have a very skewed idea about the "correct" probabilities.

Re:Microsoft and OpenAI Working on ChatGPT-Powered Bing in C... by TuringTest on Wednesday January 04, 2023 @08:41AM (#34245388)
* * *
Re:Microsoft and OpenAI Working on ChatGPT-Powered Bing in C... by ailef on Wednesday January 04, 2023 @08:57AM (#34245528)
> This means that the content that is more replicated will be considered more true by the system, regardless of its connection to reality or its coherence with the rest of the knowledge in the system.

I understand the problem, but what better way do we currently have to measure its connection to reality? At least from a practical point of view it seems that LLMs have achieved way better performance than other methods in this regard, so repeatedness doesn't look like that bad a metric. Or rather, it's the best I think we currently have.

Re:Microsoft and OpenAI Working on ChatGPT-Powered Bing in C... by mannykannot on Wednesday January 04, 2023 @09:26AM (#34245819)
> I understand the problem, but what better way do we currently have to measure its connection to reality?

We can consider its responses to a broader range of questions than those having an unambiguous and well-known answer. Its propensity for making up 'facts', and for fabricating 'explanations' that are incoherent or even self-contradictory shows that any apparent understanding of the world being represented in the text is illusory.

Re:Microsoft and OpenAI Working on ChatGPT-Powered Bing in C... by mannykannot on Wednesday January 04, 2023 @09:16AM (#34245709)
I agree, and furthermore, a search engine is constrained to pick its responses from what's already out there.

This line of thought is a distraction, anyway. The likelehood that GPT-3 will do as well as a search engine on topics where there is an unambiguous and well-known answer does little to address the more general concern.