<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Embedding on Watchstep Blog</title><link>https://blog.watchstep.site/categories/embedding/</link><description>Recent content in Embedding on Watchstep Blog</description><generator>Hugo</generator><language>en</language><copyright>Â©Â 2025 watchstep</copyright><lastBuildDate>Tue, 09 Sep 2025 17:00:00 +0900</lastBuildDate><atom:link href="https://blog.watchstep.site/categories/embedding/index.xml" rel="self" type="application/rss+xml"/><item><title>🪆 Matryoshka Representation Learning (Google, 2022) 논문 리뷰 (feat. EmbeddingGemma)</title><link>https://blog.watchstep.site/posts/mrl/</link><pubDate>Thu, 04 Sep 2025 15:10:34 +0900</pubDate><guid>https://blog.watchstep.site/posts/mrl/</guid><description>&lt;h2 id="0matryoshka-representation-learning-mrl">0/Matryoshka Representation Learning; MRL&lt;/h2>
&lt;p>gemini-embedding-001 (Google, 2025), text-embedding-3-large (2024, OpenAI), voyage-context-3 (2025, Voyage AI) 등 최신 임베딩 모델에서 Matryoshka Representation Learning; MRL를 지원하다고 하는데 MRL이 뭘까?&lt;/p></description></item></channel></rss>