Not true Gary.
This response from @openAI is true but misleading.
In the fully unscrambled cube condition, with an uninstrumented cube (ie no sensors), success rate was zero.https://twitter.com/woj_zaremba/status/1186149419739037696 …
-
-
-
basing it on bottom right corner of Table 6. am i misreading
@woj_zaremba?pic.twitter.com/F5JlXVRRig
- 1 more reply
New conversation -
-
-
I was wondering about that. Havent gotten around to reading the paper yet but does it say 0% in there? I figured given Woj's reply & since prev SOTA was 0, that he meant it was a non-zero but lower than the instrumented case.
-
New conversation -
-
-
Well technically, one could say that 0 is a result, and it is weaker.pic.twitter.com/KepFcSfaHp
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.