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.

Using GPT-3 to generate user interfaces

87 pointsby Samin100almost 5 years ago

9 comments

gas9S9zw3P9calmost 5 years ago
As someone working in the field I have a few thoughts:<p>1. The way I think about these huge models is no longer as something that makes predictions, but rather as a kind of huge knowledge base (compressed training data as representations) with smart query capabilities, which come from the fact that the model is constrained to create syntactically correct output. This could be the next generation of search. There is a fine line between search and creating novel outputs, since most novelty is just a combination of old things with constraints on correct syntax.<p>2. The shown results are likely a result of massive cherry-picking in the typical &quot;pitch deck demo&quot; fashion. You&#x27;ll see a lot more failures when trying to use this and you need to get the query &quot;just right&quot; for it to produce good output, which a skill in itself. Also let&#x27;s not forget that the model must be fed fine-tuning examples, which also need to be &quot;just right.&quot;. That&#x27;s why I believe these kinds of models aren&#x27;t that useful for fully automating things - for something to be used in production it should be 100% correct - but they are very useful as a search function and alternative to e.g. StackOverflow in this case. Querying the model will give you good enough results that you can use as a starting point for your own use case.
评论 #23857306 未加载
评论 #23857371 未加载
评论 #23857204 未加载
CJeffersonalmost 5 years ago
This is a really cool demo, and congrats to the creator for making something really cool.<p>However, while these systems based on GPT-3, and similar techniques, can produce some amazing outputs given the right inputs, you don&#x27;t have to poke them very hard before they start falling apart.<p>At the end of the day, this isn&#x27;t really any more &quot;clever&quot; than a very, very advanced and long markov chain, there isn&#x27;t any intelligence.
评论 #23857262 未加载
评论 #23857626 未加载
评论 #23857539 未加载
escardinalmost 5 years ago
What I find most interesting about this is that is has just as much trouble as humans in translating requirements into code. Admittedly it makes some fairly basic mistakes as well, but always in the under specified areas<p>i.e.:<p>You didn&#x27;t say what colour &#x27;welcome to my news letter&#x27; should be, so I picked white. The numbers from 1-5 are totally a possible random permutation in the 1-10 range. I find the crying laughing emoji to be really ugly, so it&#x27;s the worst one.
applecrazyalmost 5 years ago
This is so cool. I want to feed recipes into GPT-3 to generate novel (and humorous) food items.
jonas_kgomoalmost 5 years ago
My actual worry about this is &#x27;will I be obsolete as a front-end engineer&#x27;. It seems that AI will manage web optimization, and NLP conversational programming, so basically there is no need for a front end engineer. Perhaps I am wrong.
londons_explorealmost 5 years ago
How was this made?<p>The GPT-3 model hasn&#x27;t been published, and training it yourself would probably cost millions of dollars...
评论 #23857311 未加载
zuhayeeralmost 5 years ago
This is so cool – looking forward to using GPT-3 on random things like song lyrics and internal company slacks to see what happens
jayjagtapalmost 5 years ago
A blue color Login Button and a green color sign up button
Der_Einzigealmost 5 years ago
I was one of the people who didn&#x27;t think that transformers would be powerful enough to automate software engineers out of a job.<p>I was wrong. It&#x27;s coming for front-end engineers and it&#x27;s coming soon.