No, data annotation is not a sweatshop. (And what do you want, exactly? That the "toxic" content be left in?)
@pmddomingos
-
Reinforcement Learning Model Challenged: New Understanding Emerges
By
–
Reinforcement learning was wrong all along. Now we know why, and how to do better. https://
economist.com/science-and-te
chnology/2023/01/18/a-decades-old-model-of-animal-and-human-learning-is-under-fire
… -
Deep Learning Questions Lack Satisfying Answers
By
–
We’ve never had a satisfying answer to any question in deep learning.
-
AI Intelligence Beyond Human Cognition Models
By
–
But real AI doesn’t have to be a copy of human intelligence. (And a few years of visual input are a vast amount of data.)
-
ChatGPT Not Required for This Task, Other Apps Available
By
–
It doesn’t take ChatGPT to do that. And I’ve mentioned other killer apps before.
-
Computers as Syntactic Machines and LLMs’ Semantic Progress
By
–
Computers are syntactic machines. Large language models are a small step toward semantics.
-
Word Processors Will Soon Block Harmful Language
By
–
Soon, word processors won’t let you type harmful words.
-
Stable Diffusion: Exquisitely Rendered Nonsense
By
–
Stable diffusion: exquisitely rendered nonsense.
-
Every New Technology Creates New Forms of Spam
By
–
Every new information technology gives rise to a new kind of spam: junk mail, telemarketing, infomercials, spam email, fake reviews, Twitter bots, crap goods on Amazon, ChatGPT text, . . .