<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
<title>AI-MUSE blog</title><link href="https://ai-muse.eu/blog/"/><link rel="self" href="https://ai-muse.eu/blog/feed.xml"/><id>https://ai-muse.eu/blog/</id><updated>2026-11-09T08:00:00Z</updated>
<entry><title>Two days on Leonardo: six times fewer errors</title><link href="https://ai-muse.eu/blog/two-days-on-leonardo/"/><id>https://ai-muse.eu/blog/two-days-on-leonardo/</id><published>2026-11-07T08:00:00Z</published><updated>2026-11-09T08:00:00Z</updated><author><name>AI-MUSE</name></author><summary>Trained on one manuscript, our model makes six times fewer errors than the best general-purpose AI. Greek, Arabic, Latin and Mozart, with the numbers.</summary></entry>
<entry><title>First Time in History: An Italian Supercomputer Reads the Unreadable Voynich Manuscript with 92% Precision</title><link href="https://ai-muse.eu/blog/voynich-transcribed-not-deciphered/"/><id>https://ai-muse.eu/blog/voynich-transcribed-not-deciphered/</id><published>2026-10-11T08:00:00Z</published><updated>2026-11-07T08:00:00Z</updated><author><name>AI-MUSE</name></author><summary>On Italy’s Leonardo supercomputer at CINECA, AI-MUSE trained ten neural models to transcribe the Voynich manuscript with 91.8% character accuracy.</summary></entry>
<entry><title>The twelfth supercomputer in the world and the papers of the Resistance</title><link href="https://ai-muse.eu/blog/supercomputer-and-the-papers-of-the-resistance/"/><id>https://ai-muse.eu/blog/supercomputer-and-the-papers-of-the-resistance/</id><published>2026-08-13T08:00:00Z</published><updated>2026-11-07T08:00:00Z</updated><author><name>ELAN42</name></author><summary>Why we took a supercomputer to newspapers printed in secret in 1943-45, and why a wrong link in a Resistance archive is not just a bug.</summary></entry>
</feed>
