|
|
|
|
|
|
by Jensson
982 days ago
|
|
|
You are right, but I think it is really important to have this difference in learning in mind, because not being able to learn rules during training is the main weakness in these models currently. Understanding that weakness and how that makes their reasoning different from humans is key both to using these models and for any work on improving them. For example, you shouldn't expect it to be able to make valid chess moves reliably, that requires reading and understanding rules which it can't do during training. It can get some understanding during evaluation, but we really want to be able to encode that understanding into the model itself rather than have to keep it in eval time. |
|
|
There is a distinction between reasoning skills learned inductively (generalizing from examples), and reasoning learned deductively (via compact symbols or other structures).
The former is better at recognition of complex patterns, but can incorporate some basic deduction steps.
But explicit deduction, once it has been learned, is a far more efficient method of reasoning, and opens up our minds to vast quantities of indirect information we would never have the time or resources to experience directly.
Given how well models can do at the former, it’s going to be extremely interesting to see how quickly they exceed us at the latter - as algorithms for longer chains of processing, internal “whiteboarding” as a working memory tool for consistent reasoning over many steps and many facts, and long term retention of prompt dialogs, get developed!