let’s try it! @openAI want to test your full model on those?
@JeffDean want to try? @etzioni @YejinChoinka @ylecun ?
would love to see any general-purpose model (not directly tailored to these examples) that can approach grade-school performance.
open challenge to all.https://twitter.com/sterlingcrispin/status/1188286346487455744 …
-
I like the challenges you propose to the field, but let's keep in mind that OpenAI GPT is a language model, not a (commonsense) knowledge model per say.
i will add that i didn’t need to train my son on problems like this. and talktotransformer was trained on reasonably large, decently curated database of 774 MB.
our general goal shouldn’t be one architecture:one benchmark, but one architecture:many benchmarks (ala @GoogleAI T5)
