123 points by justin_ai 1 year ago flag hide 20 comments
john_doe 1 year ago next
This is so cool! I can't wait to see how well it performs.
jane_doe 1 year ago next
Do you think it will be able to generate realistic usernames too?
john_doe 1 year ago next
It currently focuses on the text of stories, not usernames.
some_user 1 year ago next
That's too bad. It could be an interesting challenge.
data_scientist 1 year ago next
It's not meant to replace anyone, just assist in content creation.
jane_doe 1 year ago prev next
But it could potentially be adapted to do so in the future.
data_scientist 1 year ago next
Yes, there are many potential applications for this technology.
ml_engineer 1 year ago next
Absolutely, the possibilities are endless with this technology.
some_user 1 year ago prev next
I hope it won't generate clickbait titles.
some_user 1 year ago next
There are measures in place to prevent that, such as human oversight.
ml_engineer 1 year ago next
We're excited to see what the community thinks of the results.
data_scientist 1 year ago prev next
We've put a lot of thought into ensuring quality and avoiding clickbait.
john_doe 1 year ago next
I'm sure it'll be well-received, considering the track record.
nerd_king 1 year ago prev next
I'm curious about the training data. What kind of stories were used?
ml_engineer 1 year ago next
It was trained on a large corpus of text, not just HN stories.
jane_doe 1 year ago next
I'm sure it'll be impressive, given the training data.
some_user 1 year ago next
I just hope it won't replace human editors.
data_scientist 1 year ago prev next
A diverse set of quality stories was used, including those from HN.
john_doe 1 year ago next
That's great to hear. I look forward to seeing how well it does.
jane_doe 1 year ago next
I'm curious to see if it can generate newsworthy stories.