Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
hinkley
on Aug 13, 2020
|
parent
|
context
|
favorite
| on:
Google Update Tanks Traffic
Given how often I get top results that are clearly machine-generated content, I anticipate a lot of broken overfitting will happen in the near term.
If you train NLP on text generated by NLP, you're gonna have a bad time.
colejohnson66
on Aug 13, 2020
[–]
Now I’m curious: what would happen if one
repeatedly
trained GPT-3 on text generated by a “previous generation” GPT-3? (Similar time successive JPEG saves)
AnimalMuppet
on Aug 13, 2020
|
parent
[–]
I'm pretty sure it would drift further and further from something a human would recognize as intelligible.
datameta
on Aug 13, 2020
|
root
|
parent
[–]
It would be a kind of DeepDream for NLP
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
If you train NLP on text generated by NLP, you're gonna have a bad time.