FIRP: Faster LLM inference via future intermediate representation prediction

Add code
Oct 27, 2024

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: