TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Self Supervised Learning in NLP

130 pointsby amitnessalmost 5 years ago

5 comments

why-elalmost 5 years ago
The deepemoji one is fooled by &quot;was my flight delayed? no.&quot;. I feel for the computer when it meets that one &quot;do I speak in questions?&quot; person. <i>chuckles.</i><p>On a more serious note, Hinton and other alluded to the need to restructure NLP studies as to focus more on the nature of recursion within language, which is basically what Chomsky has been saying for decades. It&#x27;s interesting to see whether they converge.
评论 #23297816 未加载
nxpnsvalmost 5 years ago
It’s like one of those low effort medium articles. But not low effort, and not on medium. I really enjoyed it.
评论 #23299491 未加载
dhabalmost 5 years ago
Thank you - I found the article helpful in giving me a lay of the land
评论 #23297815 未加载
antplsalmost 5 years ago
Hello, number 9) doesn&#x27;t say what the task is?<p>Also, I always wondered, do those methods work universally on all languages? For example Chinese, Korean and Japanese, with different alphabets.
评论 #23307811 未加载
labelbiasalmost 5 years ago
I&#x27;m wondering if these tasks have a form of bias that decreases the performance. If the model sees only positive examples and no negatives then it is biased on the positive paths of decisions. The moment where one changes the path to be incorrect, the model can&#x27;t recover from the mistake because there weren&#x27;t any negative examples during pretraining. There&#x27;s many words that never follow some words but the model never sees that.