Preaching to the choir, Greg.  See Hutter Prize.  I brought up the Turing
Test as a way of asking my question about GPT-3's "statefulness".  I took
this approach because, as you obviously are painfully aware, lots and lots
of people get excited about Turing Test style demos of GPT-3 -- so someone
may therefore be able to provide me with my answer regarding this technical
detail.

On Fri, Jun 10, 2022 at 12:13 PM Greg Staskowski <[email protected]> wrote:

> The Turing Test is useless.
>
> Period. Full stop. #NotEVenWrong.
>
> GJS
>
> On Wed, Jun 8, 2022 at 3:55 PM James Bowery <[email protected]> wrote:
>
>> Correct me if I'm wrong, but all of the attempts to use GPT-3 to pass the
>> Turing Test are "stateless" in the following sense:
>>
>> All prior inputs to GPT-3 are appended to the current input, and the
>> whole mess is sent as a single input to GPT-3 always in the same state.
>>
>> GPT-3, itself, doesn't keep track of anything, which is why all of the
>> examples of "conversations" end up being pretty limited.  You just can't
>> get away with this trick indefinitely because the input string becomes too
>> large.
>>
>> Now, not having read the Gato paper in much detail, it would _appear_
>> they are taking the "sequence" thing a bit more seriously, so there may be
>> some hope it isn't pulling the same fakery.
>>
>> Is this true?
>>
> *Artificial General Intelligence List <https://agi.topicbox.com/latest>*
> / AGI / see discussions <https://agi.topicbox.com/groups/agi> +
> participants <https://agi.topicbox.com/groups/agi/members> +
> delivery options <https://agi.topicbox.com/groups/agi/subscription>
> Permalink
> <https://agi.topicbox.com/groups/agi/T861334313bba231b-Mc46889f4e30295f71256d8da>
>

------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T861334313bba231b-M4d2ee1583cb19b0e2911f1a5
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to