Energy-Efficient LLM Serving via Disaggregated Attention--FFN and Flexible Frequency Scaling

Add code
Aug 03, 2026

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: