Boyang Yan

Home

❯

posts

❯

Which Heads Matter for Reasoning? RL-Guided KV Cache Compression

Which Heads Matter for Reasoning? RL-Guided KV Cache Compression

Jun 19, 20261 min read

reasoning model

What are Reasoning Models?

chain-of-thought (CoT)

attention head

Reference List

  1. https://arxiv.org/pdf/2510.08525

Graph View

Created with Quartz v5.0.0 © 2026