trends.barnas.net

method · first seen 27 Oct 2026

attention matching

Compacting a KV cache by selecting keys, refitting values, and adding logit biases to preserve attention outputs.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

KV cache compressionpaged attention