IAM: Efficient Inference through Attention Mapping between Different-scale LLMs

Add code
Jul 16, 2025

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: