<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>AI Research on 3rd layer</title><link>https://3rdlayer.uk/categories/ai-research/</link><description>Recent content in AI Research on 3rd layer</description><generator>Hugo -- 0.154.5</generator><language>en-US</language><lastBuildDate>Sun, 28 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://3rdlayer.uk/categories/ai-research/index.xml" rel="self" type="application/rss+xml"/><item><title>A ReLU Network Is One Giant Piecewise-Affine Function</title><link>https://3rdlayer.uk/posts/relu-piecewise-affine/</link><pubDate>Sun, 28 Jun 2026 00:00:00 +0000</pubDate><guid>https://3rdlayer.uk/posts/relu-piecewise-affine/</guid><description>A network with ReLU activations secretly carves input space into polytope pieces and, on each piece, collapses into exactly one affine map. This post builds that view up from a single neuron to interactive demos.</description></item><item><title>Sentence Reranking Using LLM Attention Maps</title><link>https://3rdlayer.uk/posts/attention-reranking/</link><pubDate>Sun, 07 Sep 2025 00:00:00 +0000</pubDate><guid>https://3rdlayer.uk/posts/attention-reranking/</guid><description>Leveraging what the LLM already knows. An experiment extracting document relevance signals from Attention Maps to rerank sentences, along with a survey of related papers.</description></item></channel></rss>