search

annargrs / blog

MIT License
0 stars 0 forks source link

2023/closed-baselines/ #10

Open utterances-bot opened 1 year ago

utterances-bot commented 1 year ago

Closed AI Models Make Bad Baselines - Hacking semantics

Will GPT-4 become a universally expected baseline in NLP research, like BERT in its time? Basic scientific methodology demands otherwise.

https://hackingsemantics.xyz/2023/closed-baselines/

Slamsneider commented 1 year ago

[META] - Would be nice with publish/update date visible in the top,

Nice article - Keep up the good work.