154 points by ignoramous 2 days ago | 16 comments
kcorbitt 2 days ago
danielhanchen 2 days ago
But all in all, the review is a reasonable "manual" I guess. I would have liked maybe more instructive comprehensive practical examples, and maybe more mention of other OSS packages for finetuning :))
YetAnotherNick 2 days ago
abc-1 2 days ago
daghamm 2 days ago
Obviously they should have stated that this is partially generated, but at least they are dog fooding it :)
worstspotgain 2 days ago
qeternity 2 days ago
The actual ORPO paper is "Odds Ratio Preference Optimisation" and it has nothing to do with pruning. This goes way beyond native language preference.
espadrine 2 days ago
cubefox 2 days ago
anothername12 2 days ago
p1esk 2 days ago
[1] https://cseweb.ucsd.edu/classes/wi08/cse253/Handouts/lecun-9...
kleiba 2 days ago
make3 2 days ago
raymond_goo 2 days ago
youoy 2 days ago
aubanel 2 days ago
The example shown in the tweet has been edited out of the paper since, but there must be others. High noise, low signal content.