Sign up or Log in
Info
This channel appears in
What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language models - Allyson Ettinger
What BERT is not: Lessons from a new su…
pdf
Challenges in Detoxifying Language Models
Challenges in Detoxifying Language Mode…
pdf
"I asked GPT-3 about Xinjiang and it broke." - Samuel Hammond 🌐🏛
"I asked GPT-3 about Xinjiang and it br…
"this effort, especially as framed in the Stanford post, implies that racist/unfair parole process could be addressed via AI (NLP/LLMs) that could “review” decisions and flag their fairness/unfairness " - Meredith Whittaker on Twitter
"this effort, especially as framed in t…
Racial Bias in Hate Speech and Abusive Language Detection Datasets - Thomas Davidson, Debasmita Bhattacharya, Ingmar Weber
Racial Bias in Hate Speech and Abusive …
pdf
"i know you don't respect my work but do you respect faculty at stanford? > all five ASR systems exhibited substantial racial disparities, with an average word error rate (WER) of 0.35 for black speakers compared with 0.19 for white speakers." - Ali Alkhatib @_alialkhatib
"i know you don't respect my work but d…
Sasha Costanza-Chock on Twitter: "Problems w/AI, #7123493478564938 Thousands of hateful trolls: "f&ck you and your transgender ideology! suck a di#k and die"toxicity data labeller: *marks 'extremely toxic, flag and ban'*me: I'm proud to be transgendertext toxicity model: 'extremely toxic, flag and ban' "
Sasha Costanza-Chock on Twitter: "Probl…
Meredith Whittaker on Twitter: "YES! And these large language models are the same ones that Stanford and its industry allies...
Meredith Whittaker on Twitter: "YES! An…
Meredith Whittaker on Twitter: "What if the power asymmetries are constitutive? What if increasing the number of people with...
Meredith Whittaker on Twitter: "What if…
An AI Helped Write This Play. It May Contain Racism
An AI Helped Write This Play. It May Co…