Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors

Add code
Oct 04, 2023

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: