<?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>E-Means on Leonardo Kuffo</title>
    <link>https://www.lkuffo.com/tags/e-means/</link>
    <description>Recent content in E-Means on Leonardo Kuffo</description>
    <image>
      <title>Leonardo Kuffo</title>
      <url>https://www.lkuffo.com/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</url>
      <link>https://www.lkuffo.com/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</link>
    </image>
    <generator>Hugo -- 0.147.7</generator>
    <language>en-us</language>
    <lastBuildDate>Tue, 17 Nov 2026 12:00:00 +0000</lastBuildDate>
    <atom:link href="https://www.lkuffo.com/tags/e-means/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>SuperEMeans: Making e-means 3× faster</title>
      <link>https://www.lkuffo.com/superemeans/</link>
      <pubDate>Tue, 17 Nov 2026 12:00:00 +0000</pubDate>
      <guid>https://www.lkuffo.com/superemeans/</guid>
      <description>&lt;blockquote&gt;
&lt;p&gt;TL;DR: We fused SuperKMeans and e-means, resulting in 3× faster clustering of high-dimensional vectors without quality trade-offs. Available now at &lt;a href=&#34;https://github.com/cwida/SuperKMeans&#34;&gt;https://github.com/cwida/SuperKMeans&lt;/a&gt;.&lt;/p&gt;&lt;/blockquote&gt;
&lt;p&gt;Recently, there has been a lot of buzz around clustering of vector embeddings: &lt;a href=&#34;https://github.com/TutteInstitute/evoc&#34;&gt;EVōC&lt;/a&gt;, &lt;a href=&#34;https://github.com/cwida/SuperKMeans&#34;&gt;SuperKMeans&lt;/a&gt;, &lt;a href=&#34;https://vectordb-ntu.github.io/RaBitQ-Library/clustering/?h=clusterin#qgkmeans&#34;&gt;QGKmeans&lt;/a&gt;, and just last week, &lt;a href=&#34;https://e-means.github.io/&#34;&gt;e-means&lt;/a&gt;, &lt;a href=&#34;https://github.com/owlgebra-ai/faiss/tree/main/contrib/o_means&#34;&gt;o-means&lt;/a&gt;, and &lt;a href=&#34;https://github.com/suayptalha/s-means&#34;&gt;s-means&lt;/a&gt;. Among these, e-means is a big deal: &lt;strong&gt;it promises the same quality as Lloyd&amp;rsquo;s with only 2 epochs&lt;/strong&gt;. Why is this a big deal for vector search?&lt;/p&gt;
&lt;p&gt;Well, clustering is a key building block for large-scale vector indexes (e.g., IVF, ScANN), which is precisely the bottleneck that users have to wait on before issuing similarity search queries.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
