Hacker News
points by nostrebored 2 days ago | 0 comments
Is it a parameter count problem? Ie why are you not seeing the cross modality bump in performance?
[–] irthomasthomas 1 day ago
Has the cross modaility bump been conclusively proven?
[–] samstevens 1 day ago
my intuition is that cross modality improvements do not really show up in practice, and especially not in small models