Hacker News new | ask | show | jobs
by turnersr 2411 days ago
What methods make BERT outdated? Do you have pointers to other options?
1 comments
bitL 2411 days ago
e.g. XLNet:

https://arxiv.org/abs/1906.08237

link
phreeza 2411 days ago
XLnet is Bert with a bunch of additional training tricks.
link
bitL 2410 days ago
BERT is a Transformer with a bunch of additional training tricks. Transformer is self-attention with a bunch of additional training tricks...
link