Discover and read the best of Twitter Threads about #benderrule
Most recents (2)
New podcast episode! 📢
@l2k and @emilymbender dive into the problems with bigger and bigger language models, the difference between form and meaning, the limits of benchmarks, and the #BenderRule.
🎥:
They discuss 4 of Emily's papers ⬇️
1/5
@l2k and @emilymbender dive into the problems with bigger and bigger language models, the difference between form and meaning, the limits of benchmarks, and the #BenderRule.
🎥:
They discuss 4 of Emily's papers ⬇️
1/5
"On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜" (Bender, Gebru et al. 2021)
Possible risks associated with bigger and bigger language models, and ways to mitigate those risks.
dl.acm.org/doi/pdf/10.114…
2/5
Possible risks associated with bigger and bigger language models, and ways to mitigate those risks.
dl.acm.org/doi/pdf/10.114…
2/5
"Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data" (Bender & Koller, 2020)
Why systems trained only on form (like language models) have no a priori way to learn meaning.
aclanthology.org/2020.acl-main.…
3/5
Why systems trained only on form (like language models) have no a priori way to learn meaning.
aclanthology.org/2020.acl-main.…
3/5
OK, I have to dust off my linguist brain for going back to work next week, so let's so how this goes. One like = one opinion about linguistics.
1. Understanding language structures, and how people use them, is relevant to so many fields. Linguistics is useful training for so many careers.
(I know this because I've interviewed lots of people about it superlinguo.com/tagged/linguis…)
(I know this because I've interviewed lots of people about it superlinguo.com/tagged/linguis…)
2. Language changes. This isn't so much an opinion, as a fact.
