<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>PPMI: Privacy-Preserving LLM Interaction with Socratic Chain-of-Thought Reasoning and Homomorphically Encrypted Vector Databases · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2506.17336/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/2506.17336/feed.xml" rel="self" type="application/rss+xml" /><item><title>PPMI: Privacy-Preserving LLM Interaction with Socratic Chain-of-Thought Reasoning and Homomorphically Encrypted Vector Databases — mentioned by cloneisme</title><link>https://news.ycombinator.com/item?id=46874261</link><guid isPermaLink="false">urn:hackerpapers:mention:2506.17336:46874261</guid><pubDate>Tue, 03 Feb 2026 17:42:11 GMT</pubDate><description>&lt;p&gt;Thanks for the interest. HEVEC uses partial homomorphic encryption, not FHE. It supports only the operations needed for an encrypted vector database (search, insert, delete, etc.), which keeps performance practical. Implementation details are in our paper: https://arxiv.org/abs/2506.17336 Happy to elaborate if helpful.&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2506.17336/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CR</category></item></channel></rss>
