contourkv

Tag

Cards List
#contourkv

Trust the Mass: Forced Weights in KV-Cache Eviction

arXiv cs.LG · 2026-08-27 Cached

This paper analyzes KV-cache eviction strategies in sparse-attention models, showing that selecting largest weights is near-optimal and that published margins come from memory and query information, with ContourKV achieving strong performance.

0 favorites 0 likes
← Back to home

Submit Feedback