3 ms·
I think the main point is — the AI don't really read and understand, they read and remember the patterns of words, which is different from understanding. They s
by bndr 4y ago
I think the main point is — the AI don't really read and understand, they read and remember the patterns of words, which is different from understanding. They see a chain of words often enough that it becomes statistically the next best output based on some input. (with some randomness in-between.)
- quonn 4y agoMaybe. Or maybe some lower layers learn to recognise language, some later layers learn to recognise concepts, some layers above learn to do some sort of reasoning. Perhaps. We have seen behaviour like that in CNNs where basic features were built up on the lower layers. The fact that a single word is output does not imply that the system is only working on that single word. Yes, that's the task and a good way - as a human - to do that task is to think about what you want to say. So the network may do something like that in between.
- akasakahakada 4y agoDo you imply that human do not learn from studying the pattern? For instance, Paragraph A and paragraph B put together, you would say that makes no sense because the logic is broken. How can you know that the logic is broken solely by those words? The only reason for meanings can never be extracted from a text is that the text contains no meanings. So now words do not contains meanings?
- reportgunner 4y agoSomething about reading between the lines I think.
- rhn_mk1 4y agoI'm tired of this trope. How is remembering patterns of words different from understanding? (Even ignoring the fact that the patterns themselves are not remembered, and LLMs don't deal with words directly but their embeddings.) Without the difference described, your words don't mean anything.
- maxdoop 4y agoAnd how does a human understand?