<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Attention on jonam'Log</title><link>https://www.jonam.io/tags/attention/</link><description>Recent content in Attention on jonam'Log</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>&amp;copy; 2026 Manoj. All Rights Reserved.</copyright><lastBuildDate>Mon, 18 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://www.jonam.io/tags/attention/index.xml" rel="self" type="application/rss+xml"/><item><title>Attention Head Similarity Pruning</title><link>https://www.jonam.io/journal/inference-engineering/research-topics/attention-head-similarity-pruning/</link><pubDate>Mon, 18 May 2026 00:00:00 +0000</pubDate><guid>https://www.jonam.io/journal/inference-engineering/research-topics/attention-head-similarity-pruning/</guid><description>Measure cross-head similarity on a prompt and skip heads that are redundant for that input.</description></item><item><title>Unlearning Layer In Attention</title><link>https://www.jonam.io/journal/inference-engineering/research-topics/unlearning-layer-in-attention/</link><pubDate>Mon, 18 May 2026 00:00:00 +0000</pubDate><guid>https://www.jonam.io/journal/inference-engineering/research-topics/unlearning-layer-in-attention/</guid><description>Can we attenuate undesirable token associations inside attention without full retraining?</description></item></channel></rss>