1<!DOCTYPE html> 2<html lang="ko"> 3<head> 4 5 <title>[ë ¼ë¬¸ 리뷰] Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning</title> 6 <meta charset="utf-8" /> 7 <meta http-equiv="X-UA-Compatible" content="IE=edge" /> 8 <meta name="HandheldFriendly" content="True" /> 9 <meta name="viewport" content="width=device-width, initial-scale=1.0" /> 10 11 <link rel="preload" as="style" href="/assets/built/screen.css?v=4f24c6eeff" /> 12 <link rel="preload" as="script" href="/assets/built/casper.js?v=4f24c6eeff" /> 13 14 <link rel="stylesheet" type="text/css" href="/assets/built/screen.css?v=4f24c6eeff" /> 15 16 <link rel="canonical" href="https://facerain.me/nonmun-ribyu-search-r1-training-llms-to-reason-and-leverage-search-engines-with-reinforcement-learning/"> 17 <meta name="referrer" content="no-referrer-when-downgrade"> 18 19 <meta property="og:site_name" content="FacerAin's blog"> 20 <meta property="og:type" content="article"> 21 <meta property="og:title" content="[ë ¼ë¬¸ 리뷰] Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning"> 22 <meta property="og:description" content="ë¤ì´ê°ë©° 23 24ì´ë² ìê°ìë LLMì´ ê²ì ìì§ê³¼ ìí¸ìì©íë©° ì¶ë¡ (Reasoning)ì ìíí ì ìë ê°í íìµ íë ììí¬ Search-R1ì ìê°í©ëë¤. ìµê·¼ OpenAIì Deep Researchë ì¬ë¬ ìµì ì°êµ¬ìì ì ì ìë¯, LLMì ì¶ë¡ ë¥ë ¥ë¿ ìëë¼ ì¤ìê° ê²ìê³¼ ê²°í©ë Reasoningì´ í° ì£¼ëª©ì ë°ê³ ììµëë¤. 25íì§ë§ 기존ì RAG(Retrieval-Augmented Generation)ì´ë Tool-Use"> 26 <meta property="og:url" content="https://facerain.me/nonmun-ribyu-search-r1-training-llms-to-reason-and-leverage-search-engines-with-reinforcement-learning/"> 27 <meta property="og:image" content="https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=2000"> 28 <meta property="article:published_time" content="2025-03-25T11:54:44.000Z"> 29 <meta property="article:modified_time" content="2026-07-07T12:50:11.000Z"> 30 <meta property="article:tag" content="Research"> 31 32 <meta property="article:publisher" content="https://www.facebook.com/ghost"> 33 <meta name="twitter:card" content="summary_large_image"> 34 <meta name="twitter:title" content="[ë ¼ë¬¸ 리뷰] Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning"> 35 <meta name="twitter:description" content="ë¤ì´ê°ë©° 36 37ì´ë² ìê°ìë LLMì´ ê²ì ìì§ê³¼ ìí¸ìì©íë©° ì¶ë¡ (Reasoning)ì ìíí ì ìë ê°í íìµ íë ììí¬ Search-R1ì ìê°í©ëë¤. ìµê·¼ OpenAIì Deep Researchë ì¬ë¬ ìµì ì°êµ¬ìì ì ì ìë¯, LLMì ì¶ë¡ ë¥ë ¥ë¿ ìëë¼ ì¤ìê° ê²ìê³¼ ê²°í©ë Reasoningì´ í° ì£¼ëª©ì ë°ê³ ììµëë¤. 38íì§ë§ 기존ì RAG(Retrieval-Augmented Generation)ì´ë Tool-Use"> 39 <meta name="twitter:url" content="https://facerain.me/nonmun-ribyu-search-r1-training-llms-to-reason-and-leverage-search-engines-with-reinforcement-learning/"> 40 <meta name="twitter:image" content="https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=2000"> 41 <meta name="twitter:label1" content="Written by"> 42 <meta name="twitter:data1" content="Yongwoo Song"> 43 <meta name="twitter:label2" content="Filed under"> 44 <meta name="twitter:data2" content="Research"> 45 <meta name="twitter:site" content="@ghost"> 46 <meta property="og:image:width" content="1200"> 47 <meta property="og:image:height" content="1799"> 48 49
49<script type="application/ld+json"> 50{ 51 "@context": "https://schema.org", 52 "@type": "Article", 53 "publisher": { 54 "@type": "Organization", 55 "name": "FacerAin's blog", 56 "url": "https://facerain.me/", 57 "logo": { 58 "@type": "ImageObject", 59 "url": "https://facerain.me/favicon.ico", 60 "width": 48, 61 "height": 48 62 } 63 }, 64 "author": { 65 "@type": "Person", 66 "name": "Yongwoo Song", 67 "url": "https://facerain.me/author/yongwoo/", 68 "sameAs": [] 69 }, 70 "headline": "[ë ¼ë¬¸ 리뷰] Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning", 71 "url": "https://facerain.me/nonmun-ribyu-search-r1-training-llms-to-reason-and-leverage-search-engines-with-reinforcement-learning/", 72 "datePublished": "2025-03-25T11:54:44.000Z", 73 "dateModified": "2026-07-07T12:50:11.000Z", 74 "image": { 75 "@type": "ImageObject", 76 "url": "https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=2000", 77 "width": 1200, 78 "height": 1799 79 },
80 "keywords": "Research", 81 "description": "ë¤ì´ê°ë©°\n\nì´ë² ìê°ìë LLMì´ ê²ì ìì§ê³¼ ìí¸ìì©íë©° ì¶ë¡ (Reasoning)ì ìíí ì ìë ê°í íìµ íë ììí¬ Search-R1ì ìê°í©ëë¤. ìµê·¼ OpenAIì Deep Researchë ì¬ë¬ ìµì ì°êµ¬ìì ì ì ìë¯, LLMì ì¶ë¡ ë¥ë ¥ë¿ ìëë¼ ì¤ìê° ê²ìê³¼ ê²°í©ë Reasoningì´ í° ì£¼ëª©ì ë°ê³ ììµëë¤.\níì§ë§ 기존ì RAG(Retrieval-Augmented Generation)ì´ë Tool-Use ë°©ìì\n\n * ë³µì¡í ë¤ë¨ê³ ê²ìì ì ì°íê² ëìí기 ì´ë µê±°ë\n * ëê·ëª¨ ê³ íì§ ì§ë íìµ ë°ì´í°ê° íìíë¤ë íê³ì ì´ ììµëë¤.\n\nì´ë° 문ì 를 í´ê²°í기 ìí´, ì´ë² ì°êµ¬ììë LLMì´ ì¤ì¤ë¡ ê²ì 쿼리를 ìì±íë©° Reasoningì ë°°ì°ë RL ê¸°ë° íë ììí¬, ë°ë¡ Search-R1ì ì ìíìµëë¤.\n\nSearch-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learn", 82 "mainEntityOfPage": "https://facerain.me/nonmun-ribyu-search-r1-training-llms-to-reason-and-leverage-search-engines-with-reinforcement-learning/" 83} 84 </script>
84 85 86 <meta name="generator" content="Ghost 6.45"> 87 <link rel="alternate" type="application/rss+xml" title="FacerAin's blog" href="https://facerain.me/rss/"> 88
88<script defer src="https://cdn.jsdelivr.net/ghost/portal@~2.69/umd/portal.min.js" data-i18n="true" data-ghost="https://facerain.me/" data-key="3fb8b569937df4745424d79246" data-api="https://facerain.me/ghost/api/content/" data-locale="ko" crossorigin="anonymous"></script>
88<style id="gh-members-styles">.gh-post-upgrade-cta-content, 89.gh-post-upgrade-cta { 90 display: flex; 91 flex-direction: column; 92 align-items: center; 93 font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, 'Open Sans', 'Helvetica Neue', sans-serif; 94 text-align: center; 95 width: 100%; 96 color: #ffffff; 97 font-size: 16px; 98} 99 100.gh-post-upgrade-cta-content { 101 border-radius: 8px; 102 padding: 40px 4vw; 103} 104 105.gh-post-upgrade-cta h2 { 106 color: #ffffff; 107 font-size: 28px; 108 letter-spacing: -0.2px; 109 margin: 0; 110 padding: 0; 111} 112 113.gh-post-upgrade-cta p { 114 margin: 20px 0 0; 115 padding: 0; 116} 117 118.gh-post-upgrade-cta small { 119 font-size: 16px; 120 letter-spacing: -0.2px; 121} 122 123.gh-post-upgrade-cta a { 124 color: #ffffff; 125 cursor: pointer; 126 font-weight: 500; 127 box-shadow: none; 128 text-decoration: underline; 129} 130 131.gh-post-upgrade-cta a:hover { 132 color: #ffffff; 133 opacity: 0.8; 134 box-shadow: none; 135 text-decoration: underline; 136} 137 138.gh-post-upgrade-cta a.gh-btn { 139 display: block; 140 background: #ffffff; 141 text-decoration: none; 142 margin: 28px 0 0; 143 padding: 8px 18px; 144 border-radius: 4px; 145 font-size: 16px; 146 font-weight: 600; 147} 148 149.gh-post-upgrade-cta a.gh-btn:hover { 150 opacity: 0.92; 151}</style> 152
152<script defer src="https://cdn.jsdelivr.net/ghost/sodo-search@~1.8/umd/sodo-search.min.js" data-key="3fb8b569937df4745424d79246" data-styles="https://cdn.jsdelivr.net/ghost/sodo-search@~1.8/umd/main.css" data-sodo-search="https://facerain.me/" data-locale="ko" crossorigin="anonymous"></script>
152 153 154 <link href="https://facerain.me/webmentions/receive/" rel="webmention"> 155
155<script defer src="/public/cards.min.js?v=4f24c6eeff"></script>
155 156 <link rel="stylesheet" type="text/css" href="/public/cards.min.css?v=4f24c6eeff"> 157
157<script defer src="/public/comment-counts.min.js?v=4f24c6eeff" data-ghost-comments-counts-api="https://facerain.me/members/api/comments/counts/"></script>
157 158
158<script defer src="/public/member-attribution.min.js?v=4f24c6eeff"></script>
158<style>:root {--ghost-accent-color: #000000;}</style> 159 <meta name="google-site-verification" content="_RKINYc3HLIqW1Nraw2D_rmdlnH8bz4itY8YZ5pysIA" /> 160<!-- Google tag (gtag.js) -->
vendor: 64 bytes, lines 160-161
160 161<script async src="https://www.googletagmanager.com/gtag/js?id=
161G-1W6H6D1TCR
vendor: 12 bytes, line 161
161"></script>
162<script> 163
vendor: 134 bytes, lines 163-167
163window.dataLayer = window.dataLayer || []; 164 function gtag(){dataLayer.push(arguments);} 165 gtag('js', new Date()); 166 167 gtag('config', '
167G-1W6H6D1TCR
vendor: 4 bytes, line 167
167');
168</script>
168 169 170<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/katex.css" crossorigin="anonymous">
171<script defer src="https://cdn.jsdelivr.net/npm/[email protected]/dist/katex.js" crossorigin="anonymous"></script>
vendor: 1 bytes, line 171
171
172<script defer src="https://cdn.jsdelivr.net/npm/[email protected]/dist/contrib/auto-render.min.js" crossorigin="anonymous" onload="renderMathInElement(document.body);"></script>
172 173 174<link href='https://rsms.me/inter/inter.css' rel='stylesheet' type='text/css'> 175<link href='https://spoqa.github.io/spoqa-han-sans/css/SpoqaHanSans-kr.css' rel='stylesheet' type='text/css'> 176<style> 177body, 178h1, h2, h3, h4, h5, h6, 179.main-nav a, 180.subscribe-button, 181.page-title, 182.post-meta, 183.read-next-story .post:before, 184.pagination, 185.site-footer, 186.post-full-content, 187.post-card-excerpt, 188.post-full-custom-excerpt, 189[class^="icon-"]:before, [class*=" icon-"]:before 190{ 191font-family: 192 "-apple-system", "BlinkMacSystemFont","Apple SD Gothic Neo", 193 "Inter", "Spoqa Han Sans", "Segoe UI", Sans-Serif, 194 "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; 195} 196</style> 197
198<script type="text/x-mathjax-config"> 199MathJax.Hub.Config({ 200 jax: ["input/TeX", "output/SVG"], 201 tex2jax: {inlineMath: [['$$', '$$']],displayMath : [['$$$$','$$$$']]}, 202}); 203</script>
vendor: 1 bytes, line 203
203
204<script src="//cdn.bootcss.com/mathjax/2.6.1/MathJax.js?config=TeX-AMS_HTML"\></script>
204 205 <link rel="preconnect" href="https://fonts.bunny.net"><link rel="stylesheet" href="https://fonts.bunny.net/css?family=noto-sans:400,700"><style>:root {--gh-font-heading: Noto Sans;--gh-font-body: Noto Sans;}</style> 206 207</head> 208<body class="post-template tag-research gh-font-heading-noto-sans gh-font-body-noto-sans is-head-middle-logo has-cover"> 209<div class="viewport"> 210 211 <header id="gh-head" class="gh-head outer"> 212 <div class="gh-head-inner inner"> 213 <div class="gh-head-brand"> 214 <a class="gh-head-logo no-image" href="https://facerain.me"> 215 FacerAin's blog 216 </a> 217 <button class="gh-search gh-icon-btn" aria-label="Search this site" data-ghost-search><svg xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2" width="20" height="20"><path stroke-linecap="round" stroke-linejoin="round" d="M21 21l-6-6m2-5a7 7 0 11-14 0 7 7 0 0114 0z"></path></svg></button> 218 <button class="gh-burger" aria-label="Main Menu"></button> 219 </div> 220 221 <nav class="gh-head-menu"> 222 <ul class="nav"> 223 <li class="nav-home"><a href="https://facerain.me/">Home</a></li> 224 <li class="nav-engineering"><a href="https://facerain.me/tag/engineering/">Engineering</a></li> 225 <li class="nav-research"><a href="https://facerain.me/tag/research/">Research</a></li> 226 <li class="nav-life"><a href="https://facerain.me/tag/life/">Life</a></li> 227 <li class="nav-github"><a href="https://github.com/FacerAin">Github</a></li> 228</ul> 229 230 </nav> 231 232 <div class="gh-head-actions"> 233 <button class="gh-search gh-icon-btn" aria-label="Search this site" data-ghost-search><svg xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke="currentColor" stroke-width="2" width="20" height="20"><path stroke-linecap="round" stroke-linejoin="round" d="M21 21l-6-6m2-5a7 7 0 11-14 0 7 7 0 0114 0z"></path></svg></button> 234 <div class="gh-head-members"> 235 <a class="gh-head-link" href="#/portal/signin" data-portal="signin">Sign in</a> 236 <a class="gh-head-button" href="#/portal/signup" data-portal="signup">Subscribe</a> 237 </div> 238 </div> 239 </div> 240 </header> 241 242 <div class="site-content"> 243 244 245 246 247<main id="site-main" class="site-main"> 248<article class="article post tag-research "> 249 250 <header class="article-header gh-canvas"> 251 252 <div class="article-tag post-card-tags"> 253 <span class="post-card-primary-tag"> 254 <a href="/tag/research/">Research</a> 255 </span> 256 </div> 257 258 <h1 class="article-title">[ë ¼ë¬¸ 리뷰] Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning</h1> 259 260 261 <div class="article-byline"> 262 <section class="article-byline-content"> 263 264 <ul class="author-list instapaper_ignore"> 265 <li class="author-list-item"> 266 <a href="/author/yongwoo/" class="author-avatar author-profile-image" aria-label="Read more of Yongwoo Song"><svg viewBox="0 0 24 24" xmlns="http://www.w3.org/2000/svg"><g fill="none" fill-rule="evenodd"><path d="M3.513 18.998C4.749 15.504 8.082 13 12 13s7.251 2.504 8.487 5.998C18.47 21.442 15.417 23 12 23s-6.47-1.558-8.487-4.002zM12 12c2.21 0 4-2.79 4-5s-1.79-4-4-4-4 1.79-4 4 1.79 5 4 5z" fill="#FFF"/></g></svg> 267</a> 268 </li> 269 </ul> 270 271 <div class="article-byline-meta"> 272 <h4 class="author-name"><a href="/author/yongwoo/">Yongwoo Song</a></h4> 273 <div class="byline-meta-content"> 274 <time class="byline-meta-date" datetime="2025-03-25">25 3ì 2025</time> 275 <span class="byline-reading-time"><span class="bull">•</span> 11 min read</span> 276 </div> 277 </div> 278 279 </section> 280 <a href="#/share" class="gh-button gh-button-share">Share</a> 281 </div> 282 283 <figure class="article-image"> 284 <img 285 srcset="https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=300 300w, 286 https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=600 600w, 287 https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=1000 1000w, 288 https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=2000 2000w" 289 sizes="(min-width: 1400px) 1400px, 92vw" 290 src="https://images.unsplash.com/photo-1742325989789-b42912a531dd?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDN8fHx8fHx8fDE3NDI4ODc1Nzl8&ixlib=rb-4.0.3&q=80&w=2000" 291 alt="[ë ¼ë¬¸ 리뷰] Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning" 292 fetchpriority="high" 293 loading="eager" 294 />
295 <figcaption><span style="white-space: pre-wrap;">Photo by </span><a href="https://unsplash.com/@steelhouseproductions?utm_source=ghost&utm_medium=referral&utm_campaign=api-credit"><span style="white-space: pre-wrap;">Tom Nagel</span></a><span style="white-space: pre-wrap;"> / </span><a href="https://unsplash.com/?utm_source=ghost&utm_medium=referral&utm_campaign=api-credit"><span style="white-space: pre-wrap;">Unsplash</span></a></figcaption> 296 </figure> 297 298 </header> 299 300 <section class="gh-content gh-canvas"> 301 <h3 id="%EB%93%A4%EC%96%B4%EA%B0%80%EB%A9%B0">ë¤ì´ê°ë©°</h3><p>ì´ë² ìê°ìë LLMì´ ê²ì ìì§ê³¼ ìí¸ìì©íë©° ì¶ë¡ (Reasoning)ì ìíí ì ìë ê°í íìµ íë ììí¬ <strong>Search-R1</strong>ì ìê°í©ëë¤. ìµê·¼ OpenAIì Deep Researchë ì¬ë¬ ìµì ì°êµ¬ìì ì ì ìë¯, LLMì ì¶ë¡ ë¥ë ¥ë¿ ìëë¼ <strong>ì¤ìê° ê²ìê³¼ ê²°í©ë Reasoning</strong>ì´ í° ì£¼ëª©ì ë°ê³ ììµëë¤.<br>íì§ë§ 기존ì <strong>RAG(Retrieval-Augmented Generation)</strong>ì´ë <strong>Tool-Use</strong> ë°©ìì</p><ul><li>ë³µì¡í ë¤ë¨ê³ ê²ìì ì ì°íê² ëìí기 ì´ë µê±°ë</li><li>ëê·ëª¨ ê³ íì§ ì§ë íìµ ë°ì´í°ê° íìíë¤ë íê³ì ì´ ììµëë¤.</li></ul><p>ì´ë° 문ì 를 í´ê²°í기 ìí´, ì´ë² ì°êµ¬ììë <strong>LLMì´ ì¤ì¤ë¡ ê²ì 쿼리를 ìì±íë©° Reasoningì ë°°ì°ë RL ê¸°ë° íë ììí¬</strong>, ë°ë¡ <strong>Search-R1</strong>ì ì ìíìµëë¤.</p><figure class="kg-card kg-bookmark-card"><a class="kg-bookmark-container" href="https://arxiv.org/abs/2503.09516?ref=facerain.me"><div class="kg-bookmark-content"><div class="kg-bookmark-title">Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning</div><div class="kg-bookmark-description">Efficiently acquiring external knowledge and up-to-date information is essential for effective reasoning and text generation in large language models (LLMs). Prompting advanced LLMs with reasoning capabilities during inference to use search engines is not optimal, since the LLM does not learn how to optimally interact with the search engine. This paper introduces Search-R1, an extension of the DeepSeek-R1 model where the LLM learns -- solely through reinforcement learning (RL) -- to autonomously generate (multiple) search queries during step-by-step reasoning with real-time retrieval. Search-R1 optimizes LLM rollouts with multi-turn search interactions, leveraging retrieved token masking for stable RL training and a simple outcome-based reward function. Experiments on seven question-answering datasets show that Search-R1 improves performance by 26% (Qwen2.5-7B), 21% (Qwen2.5-3B), and 10% (LLaMA3.2-3B) over strong baselines. This paper further provides empirical insights into RL optimization methods, LLM choices, and response length dynamics in retrieval-augmented reasoning. The code and model checkpoints are available at https://github.com/PeterGriffinJin/Search-R1.</div><div class="kg-bookmark-metadata"><img class="kg-bookmark-icon" src="https://arxiv.org/static/browse/0.3.4/images/icons/apple-touch-icon.png" alt=""><span class="kg-bookmark-author">arXiv.org</span><span class="kg-bookmark-publisher">Bowen Jin</span></div></div><div class="kg-bookmark-thumbnail"><img src="https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png" alt=""></div></a></figure><figure class="kg-card kg-bookmark-card"><a class="kg-bookmark-container" href="https://github.com/PeterGriffinJin/Search-R1?ref=facerain.me"><div class="kg-bookmark-content"><div class="kg-bookmark-title">GitHub - PeterGriffinJin/Search-R1: Search-R1: An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL</div><div class="kg-bookmark-description">Search-R1: An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL - PeterGriffinJin/Search-R1</div><div class="kg-bookmark-metadata"><img class="kg-bookmark-icon" src="https://github.githubassets.com/assets/pinned-octocat-093da3e6fa40.svg" alt=""><span class="kg-bookmark-author">GitHub</span><span class="kg-bookmark-publisher">PeterGriffinJin</span></div></div>
301<div class="kg-bookmark-thumbnail"><img src="https://opengraph.githubassets.com/e06fce9804724dd3173758bc21e7213f33c70fd1876c67c072e0c9af1b30604f/PeterGriffinJin/Search-R1" alt=""></div></a></figure><h3 id="abstract-conclusion">Abstract & Conclusion</h3><ul><li>기존 ë°©ë²ë¡ (RAG, Tool-Use)ì <strong>LLMì´ ê²ì ìì§ê³¼ ìí¸ìì©íë ë²</strong>ì ë°°ì°ì§ 못í¨.</li><li>Search-R1ì <strong>ê°í íìµ ê¸°ë°ì¼ë¡ LLMì´ ë¤ë¨ê³ë¡ ê²ì 쿼리를 ìì±íë©´ì Reasoning</strong>ì ìííëë¡ íë ¨.</li><li>7ê°ì QA ë°ì´í°ì
íê°ìì í° ì±ë¥ í¥ì:Qwen2.5-7B: <strong>26%</strong>Qwen2.5-3B: <strong>21%</strong>LLaMA3.2-3B: <strong>10%</strong></li></ul><h3 id="introduction">Introduction</h3><p>LLMì´ ì¸ë¶ ì§ìì íì©í기 ìí´ìë RAGì Tool-Use í¬ê² ë ê°ì§ ë°©ë²ì´ ììµëë¤. íì§ë§ ê° ë°©ë²ì ìëì ê°ì íê³ê° ììµëë¤.</p><p><strong>RAG (Retrieval-Augmented Generation)</strong> - ì
ë ¥ 쿼리를 ë°íì¼ë¡ ê´ë ¨ Passage를 ê²ìíì¬ Contextì í¨ê» LLMì Inputì¼ë¡ ì ê³µ</p><ul><li>ê²ìë ì ë³´ê° ê´ë ¨ì±ì´ ìì ìë ìì</li><li>Multi-hop retrieval ì§ìíì§ ìì</li><li>LLMì´ Search Engineê³¼ ìí¸ìì© íë ë°©ë²ì ë°°ì°ì§ 못í¨</li></ul><p><strong>Search Engine Tool-Use - </strong>Search Engineì Toolë¡ íì©íëë¡ íìµ</p><ul><li>ì¼ë°íì ì´ë ¤ì</li><li>íì¥ì± 문ì : íë ¨ ê¸°ë° ë°©ìì ëì íì§ì ëê·ëª¨ annotated trajectoriesê° íìí¨</li></ul><p></p><p>ë°ë¼ì ì ìë¤ì ì´ë¬í 문ì 를 í´ê²°í기 ìí´ ê°ííìµ ê¸°ë° ë°©ë²ë¡ ì¸ S<strong>earch-R1ì ì ì</strong>í©ëë¤.</p><ul><li>íìµ ì Search Engineê³¼ì ìí¸ìì©ì íìµì ì¼ë¶ë¡ í¬í¨</li><li>Multi-Turn Interleaved Reasoning and Search</li><li>Reward Design</li></ul><h3 id="methodology">Methodology</h3><p>ê·¸ë¼ ì°êµ¬íì RLì ë°íì¼ë¡ ì´ë»ê² Search Engineê³¼ì ìí¸ìì©ì íìµ ìí¬ ì ìììê¹ì?</p><p>Search-R1ì Objectiveë ë¤ìê³¼ ê°ìµëë¤.</p><p>$$\max_{\pi} E_{x \sim D, y \sim \pi_{\theta}(\cdot | x; R)} \left[ r \phi(x, y)\right] - \beta D_{KL} \left[ \pi_{\theta}(y | x; R) | \pi_{ref}(y | x; R) \right]$$</p><p>ì¼ë°ì ì¸ RL Objectiveì í¬ê² ë¤ë¥´ì§ ììµëë¤.</p><p>$$\pi_\theta(\cdot | x; R)$$</p><p>Rì Search Engineì ì미í©ëë¤. Policy LLMì R (Search Engine)ê³¼ í¨ê» Reward를 ìµëíí ì ìëë¡ íìµë©ëë¤. ì´ë¥¼ íµí´ LLMì´ Search Engineê³¼ í¨ê³¼ì ì¼ë¡ Interact íë ë°©ë²ì ë°°ì¸ ì ìë ê²ì´ì£ .</p><p>êµì¥í ëì´ë¸íë©´ìë ê°ë¨í ì ê·¼ ë°©ìì
ëë¤. ð¤</p><h4 id="ppo-search-engine">PPO + Search Engine</h4><p>ìì ì ìí RL objectiveë PPO í¹ì GRPOë¡ íìµ (Policy Optimization) í ì ìë¤ê³ í©ëë¤. ìì ë¤ìê³¼ ê°ìµëë¤.</p><p>\[J_{PPO}(\theta) = E_{x \sim D, y \sim \pi_{old}( \cdot | x; R)} \left[ \frac{1}{\sum_{t=1}^{|y|} I(y_t)} \sum_{t=1}^{|y|} I(y_t) \cdot \min \left( \frac{\pi_{\theta}(y_t | x, y_{<t}; R)}{\pi_{old}(y_t | x, y_{<t}; R)} A_t,\ \text{clip} \left( \frac{\pi_{\theta}(y_t | x, y_{<t}; R)}{\pi_{old}(y_t | x, y_{<t}; R)} , 1 - \epsilon,\ 1 + \epsilon \right) A_t \right) \right]\]</p><p>Policy Modelìì Search Engine Rì í¨ê» ì¬ì©íë¤ë ê²ì ì ì¸íë©´ ì¼ë°ì ì¸ PPOì ìê³¼ í¬ê² ë¤ë¥´ì§ ììµëë¤.</p><p>í ê°ì§ 주목í ì ì <strong>ê²ìë í í°ì ëí´ìë Loss Maskingì ì§í</strong>íë¤ë ì ì
ëë¤.</p><p>$I(y_t)$ ìì<strong> LLMì ìí´ ìì±ë í í°ì¸ ê²½ì°, loss를 ê³ì°íê³ íìµì ë°ì</strong>íë íí¸, <strong>Search Engineìì ê²ìë í í°ì¸ ê²½ì° loss ê³ì°ìì ì ì¸</strong>í©ëë¤.</p><p>ì´ë¥¼ íµí´ LLMì´ ì§ì ìì±í ê²ì´ ìë í í°ì íìµ ê³¼ì ìì ì ì¸íì¬ <strong>ë¶íìí íìµì ë°©ì§íê³ íìµ ìì ì±ì ëì¼ ì ììë¤</strong>ê³ í©ëë¤.</p><p></p><h4 id="grpo-search-engine">GRPO + Search Engine</h4><p>ìµê·¼ Deepseekë¡ í«í GRPOë íì©ì´ ê°ë¥í©ëë¤.</p><figure class="kg-card kg-image-card"><img src="https://facerain.me/content/images/2025/03/image-6.png" class="kg-image" alt="" loading="lazy" width="1388" height="287" srcset="https://facerain.me/content/images/size/w600/2025/03/image-6.png 600w, https://facerain.me/content/images/size/w1000/2025/03/image-6.png 1000w, https://facerain.me/content/images/2025/03/image-6.png 1388w" sizes="(min-width: 720px) 720px"></figure><p></p><p>PPOì ë§ì°¬ê°ì§ë¡ 기존 GRPO ìê³¼ í¬ê² ë¤ë¥´ì§ ìì¼ë©°, ê²ìë í í°ì ëí´ìë Loss Maskingì ì§ííììµëë¤.</p><h4 id="text-generation-with-interleaved-multi-turn-search-engine-call">Text Generation with Interleaved Multi-turn Search Engine Call</h4><p>ê·¸ë ë¤ë©´ LLMì ì´ë»ê² ìì± (ì¶ë¡ )ì ì§ííë ëì Search Engineê³¼ ìí¸ìì©ì í ì ìììê¹ì?</p><p>모ë¸ì 3ê°ì§ Special Tokenì íì©í©ëë¤.</p><p><strong><search></search></strong>: í´ë¹ í í° ë´ë¶ì í
ì¤í¸ë ê²ì ì¿¼ë¦¬ë¡ íì±íì¬ Search Engineì í¸ì¶</p><p><strong><information></information></strong>
301: ìì Search Engineì ìí ê²ì ê²°ê³¼</p><p><strong><answer></answer></strong>: ìµì¢ ìëµ</p><p>ê·¸ë¦¬ê³ íìµì ìí´ ì¬ì©í Trainiing Templateì ìëì ê°ë¤ê³ í©ëë¤.</p><p>ìì ì´í´ë³¸ think, search, answer í í°ì´ íì©ëê³ ììµëë¤.</p><pre><code>Answer the given question. You must conduct reasoning inside <think> and </think> 302first every time you get new information. After reasoning, if you find you lack some 303knowledge, you can call a search engine by <search> query </search>, and it will 304return the top searched results between <information> and </information>. You 305can search as many times as you want. If you find no further external knowledge 306needed, you can directly provide the answer inside <answer> and </answer> without 307detailed illustrations. For example, <answer> Beijing </answer>. Question: question.</code></pre><h4 id=""></h4><h3 id="reward-modeling">Reward Modeling</h3><p>본 íë ììí¬ìì Reward Modelì ì´ë»ê² ì¤ê³íììê¹ì?</p><p>$$r_\phi(x, y) = EM(a_{pred}, a_{gold})$$</p><p>ì°êµ¬íì 모ë¸ì´ ì측í ëµë³ê³¼ ì ëµ ì¬ì´ì<strong> EM (Exact Match) Score</strong>를 ì¬ì©íììµëë¤. ëíì ì¸ ë¤ë¥¸ Reward Model ë°©ë² ì¤ íëì¸ Neural Reward Modelì Reward Hacking (모ë¸ì´ ì¤íë ¤ í¸ë²ì¼ë¡ Rewardë§ ëì´ë ¤ í´ì ì ìê° í¥ìëì§ ìë íì)ê³¼ ê³ì° ë¹ì© ë° ë³µì¡ì± 문ì ê° ììë¤ê³ í©ëë¤.</p><p>ëí Deepseek-R1ìì íì©í format rewardë 모ë¸ì´ ì´ë¯¸ 구조ì ê·ì¹ì ì ë°ë¥´ê¸° ë문ì ë³ëë¡ ì ì©íì§ ììë¤ê³ í©ëë¤.</p><p></p><h3 id="main-results">Main Results</h3><p>ì°êµ¬íì íê°ë¥¼ ìí´ 7ê°ì§ QA ë°ì´í°ì ì íì©íì¬ íê° ë²¤ì¹ë§í¬ë¥¼ 구ì±íììµëë¤. (General QA: NQ, TriviaQA / Multi-Hop Question Answering: HotpotQA, 2WikiMultiHopQA, Musique, Banboogle)</p><p>ë¹êµ íê°ë¥¼ ìí´ 3ê°ì§ ë¹êµêµ°ì ì±ííììµëë¤.</p><ul><li>Inference without Retrieval<ul><li>External Knoledge ìì´ Direct Inference í¹ì CoT ì¶ë¡ </li></ul></li><li>Inference with Retrieval<ul><li>RAG, IRCoT, Search-o1</li></ul></li><li>Fine-Tuning-Based Methods<ul><li>SFT</li><li>R1 (without a Search Engine)</li></ul></li></ul><p>ê·¸ë¦¬ê³ Qwen-2.5-3B(Base/Instruct), Qwen-2.5-7B(Base/Instruct), Llama-3.2-3B(Base/Instruct)를 íì©íì¬ íìµì ì§ííìê³ ,</p><p>Retrieverë¡ë E5를 íì©íì¬ Wikipedia dump를 ì ê³µíë¤ê³ í©ëë¤.</p><figure class="kg-card kg-image-card"><img src="https://facerain.me/content/images/2025/03/image-1.png" class="kg-image" alt="" loading="lazy" width="554" height="584"></figure><p>ê²°ê³¼ë ìì ê°ìµëë¤.</p><p>ëë¶ë¶ì ë°ì´í°ì ìì ëì ì±ë¥ì ë³´ì´ê³ ììµëë¤.</p><p></p><h3 id="analysis">Analysis</h3><p>ê°ì¸ì ì¼ë¡ í¥ë¯¸ë¡ê² ì½ìë ë¶ë¶ì ëë¤!</p><p>ì°êµ¬íì Search-R1 Methodìì ì¬ë¬ê°ì§ ë¹êµ ì¤íì ì§ííìëë°ì,</p><p><strong>GRPO vs PPO</strong></p><ul><li>GRPOë PPOë³´ë¤ ë¹ ë¥´ê² ìë ´, PPOë critic modelì ì¬ì©í기 ìí ì´ê¸° íìµìì ì¤ë¹ ë¨ê³ê° íìí¨</li><li>PPOë GRPOë³´ë¤ íë ¨ ìì ì±ì´ ë ëì</li><li>PPOì GRPOì ìµì¢ íë ¨ ë³´ìì ë¹ì·í¨</li><li>ëª¨ë¸ ë³ë¡ë ë¤ë¥¸ ììì ë³´ì¼ ëë ìì</li></ul><figure class="kg-card kg-image-card"><img src="https://facerain.me/content/images/2025/03/image-2.png" class="kg-image" alt="" loading="lazy" width="656" height="124" srcset="https://facerain.me/content/images/size/w600/2025/03/image-2.png 600w, https://facerain.me/content/images/2025/03/image-2.png 656w"></figure><p><strong>Base vs Instruct</strong></p><ul><li>Instruct 모ë¸ì´ ì´ê¸° ì±ë¥ë ëê³ ìë ´ ìëë ë ë¹ ë¦</li><li>íì§ë§ íë ¨ì´ ì§íë¨ì ë°ë¼ Base 모ë¸ì Instruct 모ë¸ê³¼ ë¹ì·í ìì¤ì ì±ë¥ì ëë¬í ì ìì</li><li>RLì´ ì¼ë°ì ì¸ SFTì Pre-training ê°ì 격차를 í´ìí ì ìë¤ê³ 주ì¥</li></ul><figure class="kg-card kg-image-card"><img src="https://facerain.me/content/images/2025/03/image-3.png" class="kg-image" alt="" loading="lazy" width="657" height="156" srcset="https://facerain.me/content/images/size/w600/2025/03/image-3.png 600w, https://facerain.me/content/images/2025/03/image-3.png 657w"></figure><p><strong>Response Length Study</strong></p><ul><li>LLaMA3.2-3b-base ëª¨ë¸ ê¸°ë°, NQ ë°ì´í°ì ì¼ë¡ íë ¨</li><li>ì´ê¸° ë¨ê³ (~100 ì¤í )<ul><li>ìëµ ê¸¸ì´ê° ê¸ê²©í ê°ì</li><li>íë ¨ rewardì ì½ê° ì¦ê°</li><li>모ë¸ì´ ë¶íìí ë¨ì´ë¥¼ ì ê±°íê³ ìì ì구 ì¬íì ì ì</li></ul></li><li>ì¤ê° ë¨ê³ (100-130 ì¤í )<ul><li>ìëµ ê¸¸ì´ì íë ¨ ë³´ìì´ ëª¨ë í¬ê² ì¦ê°</li><li>LLMì´ Search Engineì í¸ì¶íë ê²ì íìµíì¬ ê²ì ê²°ê
307³¼ë¡ ì¸í ìëµì´ 길ì´ì§</li><li>모ë¸ì´ ê²ì 결과를 í¨ê³¼ì ì¼ë¡ íì©íì¬ íë ¨ rewardê° í¬ê² í¥ì</li></ul></li><li>í기 ë¨ê³ (130 ì¤í ì´í)<ul><li>ìëµ ê¸¸ì´ê° ìì í</li><li>íë ¨ Rewardë ê³ì ì½ê° ì¦ê°</li><li>모ë¸ì´ Search Engineì í¨ê³¼ì ì¼ë¡ ì¬ì©íë ë°©ë²ì íìµíê³ Search Query를 ê°ì íëë° ì§ì¤</li></ul></li></ul><figure class="kg-card kg-image-card"><img src="https://facerain.me/content/images/2025/03/image-4.png" class="kg-image" alt="" loading="lazy" width="984" height="305" srcset="https://facerain.me/content/images/size/w600/2025/03/image-4.png 600w, https://facerain.me/content/images/2025/03/image-4.png 984w" sizes="(min-width: 720px) 720px"></figure><p><strong>Study of Retrieved Tokens Loss Masking</strong></p><ul><li>Loss Maskingì ì ì©íë©´ íìµì´ ëì± ìì ì ì¼ë¡ ì§íëê³ , ì ë°ì ì¸ ì±ë¥ì´ í¥ìë¨.</li></ul><figure class="kg-card kg-image-card"><img src="https://facerain.me/content/images/2025/03/image-5.png" class="kg-image" alt="" loading="lazy" width="548" height="68"></figure><h3 id="%EB%A7%88%EC%B9%98%EB%A9%B0">ë§ì¹ë©°</h3><p>Search-R1ì ë¨ìí RAGë í´ ì¬ì©ì ëì´ì, <strong>LLMì´ ì¤ì¤ë¡ Reasoning ê³¼ì ììì ê²ìì ì ëµì ì¼ë¡ íì©íëë¡ RL ì ëµì¼ë¡ íìµìí¨</strong> ì ì´ ì¸ì ê¹ìëë°ì,<br>ìµê·¼ LLMì Reasoning ë° Agent ì°êµ¬ë¤ì´ íë°í´ì§ê³ ìëë°, <strong>Search-R1</strong>ì ì´ íë¦ ììì Search Engineì LLMì ì´ë»ê² ë ¹ì¬ë¼ ì ììì§ í¥ë¯¸ë¡ì´ ê´ì ì ì ìíë¤ê³ ìê°í©ëë¤.</p><p>ê°ì¬í©ëë¤ ð</p> 308 </section> 309 310 <section class="article-comments gh-canvas"> 311 312
312<script defer src="https://cdn.jsdelivr.net/ghost/comments-ui@~1.5/umd/comments-ui.min.js" data-locale="ko" data-ghost-comments="https://facerain.me/" data-api="https://facerain.me/ghost/api/content/" data-admin="https://facerain.me/ghost/" data-key="3fb8b569937df4745424d79246" data-title="null" data-count="true" data-post-id="67e25a922a4be6054c3df8ef" data-color-scheme="auto" data-avatar-saturation="60" data-accent-color="#000000" data-comments-enabled="all" data-publication="FacerAin's blog" crossorigin="anonymous"></script>
312 313 314 </section> 315 316</article> 317</main> 318 319 320 321 322 <aside class="read-more-wrap outer"> 323 <div class="read-more inner"> 324 325<article class="post-card post"> 326 327 <a class="post-card-image-link" href="/dogseo-sagooeju/"> 328 329 <img class="post-card-image" 330 srcset="https://images.unsplash.com/photo-1789921464199-d87f82e3f8d2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDExfHx8fHx8fHwxNzkxMDM3MzIxfA&ixlib=rb-4.1.0&q=80&w=300 300w, 331 https://images.unsplash.com/photo-1789921464199-d87f82e3f8d2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDExfHx8fHx8fHwxNzkxMDM3MzIxfA&ixlib=rb-4.1.0&q=80&w=600 600w, 332 https://images.unsplash.com/photo-1789921464199-d87f82e3f8d2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDExfHx8fHx8fHwxNzkxMDM3MzIxfA&ixlib=rb-4.1.0&q=80&w=1000 1000w, 333 https://images.unsplash.com/photo-1789921464199-d87f82e3f8d2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDExfHx8fHx8fHwxNzkxMDM3MzIxfA&ixlib=rb-4.1.0&q=80&w=2000 2000w" 334 sizes="(max-width: 1000px) 400px, 800px" 335 src="https://images.unsplash.com/photo-1789921464199-d87f82e3f8d2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDExfHx8fHx8fHwxNzkxMDM3MzIxfA&ixlib=rb-4.1.0&q=80&w=600" 336 alt="[ë ì] ì¬ê³ ì¸ì£¼" 337 loading="lazy" 338 /> 339 340 341 </a> 342 343 <div class="post-card-content"> 344 345 <a class="post-card-content-link" href="/dogseo-sagooeju/"> 346 <header class="post-card-header"> 347 <div class="post-card-tags"> 348 </div> 349 <h2 class="post-card-title"> 350 [ë ì] ì¬ê³ ì¸ì£¼ 351 </h2> 352 </header> 353 <div class="post-card-excerpt">ì¬ê³ ì¸ì£¼ | íì§ê¸° | ì´í¬ë¡ì¤ - ìì¤24âì°ë¦¬ë ì§ê¸ âëëí 무ë¥í¨âì ë¹ ì ¸ ìë¤âê³¼ì ìì´ ê²°ë¡ ë§ ì»ë ìë, ì¸ê°ì 무ìì ìê² ëëê°?ì°ì¸ë íì§ê¸° êµì, AI ìë ê°ì¥ ì¡°ì©íê³ ìíí ë³í를 ë§íë¤ë§¤ëë¬ì´ ë¬¸ì¥ ì¬ì´ë¡, ì´ë ë ë¶í° íìì ì¼êµ´ì´ ë³´ì´ì§ ì기 ììíë¤. ì°ì¸ë íê³µìëª ê³µíâ¦ì´í¬ë¡ì¤íì§ê¸° 354 * íì¤í: AIìê² ì¬ì 를 ë¹¼ì겨ë²ë¦° ì°ë¦¬ìê² ê±´ë¤ë ìë´ì (5/</div> 355 </a> 356 357 <footer class="post-card-meta"> 358 <time class="post-card-meta-date" datetime="2026-10-04">04 10ì 2026</time> 359 <span class="post-card-meta-length">8 min read</span> 360
360<script 361 data-ghost-comment-count="6ac11d6abcedbc0001dc661d" 362 data-ghost-comment-count-empty="" 363 data-ghost-comment-count-singular="comment" 364 data-ghost-comment-count-plural="comments" 365 data-ghost-comment-count-tag="span" 366 data-ghost-comment-count-class-name="" 367 data-ghost-comment-count-autowrap="true" 368> 369</script>
369 370 </footer> 371 372 </div> 373 374</article> 375 376<article class="post-card post"> 377 378 <a class="post-card-image-link" href="/beonyeog-the-second-half/"> 379 380 <img class="post-card-image" 381 srcset="https://images.unsplash.com/photo-1760592150404-adacb88548e2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDJ8fHx8fHx8fDE3NjA2NTUzOTJ8&ixlib=rb-4.1.0&q=80&w=300 300w, 382 https://images.unsplash.com/photo-1760592150404-adacb88548e2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDJ8fHx8fHx8fDE3NjA2NTUzOTJ8&ixlib=rb-4.1.0&q=80&w=600 600w, 383 https://images.unsplash.com/photo-1760592150404-adacb88548e2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDJ8fHx8fHx8fDE3NjA2NTUzOTJ8&ixlib=rb-4.1.0&q=80&w=1000 1000w, 384 https://images.unsplash.com/photo-1760592150404-adacb88548e2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDJ8fHx8fHx8fDE3NjA2NTUzOTJ8&ixlib=rb-4.1.0&q=80&w=2000 2000w" 385 sizes="(max-width: 1000px) 400px, 800px" 386 src="https://images.unsplash.com/photo-1760592150404-adacb88548e2?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDJ8fHx8fHx8fDE3NjA2NTUzOTJ8&ixlib=rb-4.1.0&q=80&w=600" 387 alt="[ë²ì] The Second Half" 388 loading="lazy" 389 /> 390 391 392 </a> 393 394 <div class="post-card-content"> 395 396 <a class="post-card-content-link" href="/beonyeog-the-second-half/"> 397 <header class="post-card-header"> 398 <div class="post-card-tags"> 399 </div> 400 <h2 class="post-card-title"> 401 [ë²ì] The Second Half 402 </h2> 403 </header> 404 <div class="post-card-excerpt">The Second Halftldr: Weâre at AIâs halftime.Shunyu Yao 405 406ì´ë² ìê°ìë Shunya Yaoì The Second Halfì í¬ì¤í¸ë¥¼ ìê°í©ëë¤. ìµê·¼ Deepseek R1, OpenAI O-Series ë± ê°ííìµì ì ì©í Reasoning 모ë¸ë¤ì´ ëì ì±ë¥ì¼ë¡ ëê°ì ë³´ì´ê³ ììµëë¤. 무ìì´ ì¸ì´ 모ë¸ì ê°ííìµì ì±ê³µíê² íìê¹ì? ëí 2025ë 4ì ê¸°ì¤ AIë ì§ê¸ê¹ì§ ì´ë¤ 길ì</div> 407 </a> 408 409 <footer class="post-card-meta"> 410 <time class="post-card-meta-date" datetime="2025-10-20">20 10ì 2025</time> 411 <span class="post-card-meta-length">23 min read</span> 412
412<script 413 data-ghost-comment-count="68f17584237318cbc3a0ec08" 414 data-ghost-comment-count-empty="" 415 data-ghost-comment-count-singular="comment" 416 data-ghost-comment-count-plural="comments" 417 data-ghost-comment-count-tag="span" 418 data-ghost-comment-count-class-name="" 419 data-ghost-comment-count-autowrap="true" 420> 421</script>
421 422 </footer> 423 424 </div> 425 426</article> 427 428<article class="post-card post"> 429 430 <a class="post-card-image-link" href="/paper-review-colpali-efficient-document-retrieval-with-vision-language-models/"> 431 432 <img class="post-card-image" 433 srcset="https://images.unsplash.com/photo-1499428665502-503f6c608263?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDcyfHx8fHx8fHwxNzUyNDgwODc3fA&ixlib=rb-4.1.0&q=80&w=300 300w, 434 https://images.unsplash.com/photo-1499428665502-503f6c608263?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDcyfHx8fHx8fHwxNzUyNDgwODc3fA&ixlib=rb-4.1.0&q=80&w=600 600w, 435 https://images.unsplash.com/photo-1499428665502-503f6c608263?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDcyfHx8fHx8fHwxNzUyNDgwODc3fA&ixlib=rb-4.1.0&q=80&w=1000 1000w, 436 https://images.unsplash.com/photo-1499428665502-503f6c608263?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDcyfHx8fHx8fHwxNzUyNDgwODc3fA&ixlib=rb-4.1.0&q=80&w=2000 2000w" 437 sizes="(max-width: 1000px) 400px, 800px" 438 src="https://images.unsplash.com/photo-1499428665502-503f6c608263?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8YWxsfDcyfHx8fHx8fHwxNzUyNDgwODc3fA&ixlib=rb-4.1.0&q=80&w=600" 439 alt="[ë ¼ë¬¸ 리뷰] ColPali: EFFICIENT DOCUMENT RETRIEVAL WITH VISION LANGUAGE MODELS" 440 loading="lazy" 441 /> 442 443 444 </a> 445 446 <div class="post-card-content"> 447 448 <a class="post-card-content-link" href="/paper-review-colpali-efficient-document-retrieval-with-vision-language-models/"> 449 <header class="post-card-header"> 450 <div class="post-card-tags"> 451 </div> 452 <h2 class="post-card-title"> 453 [ë ¼ë¬¸ 리뷰] ColPali: EFFICIENT DOCUMENT RETRIEVAL WITH VISION LANGUAGE MODELS 454 </h2> 455 </header> 456 <div class="post-card-excerpt">ColPali: Efficient Document Retrieval with Vision Language ModelsDocuments are visually rich structures that convey information through text, but also figures, page layouts, tables, or even fonts. Since modern retrieval systems mainly rely on the textual information they extract from document pages to index documents -often through lengthy and brittle processes-</div> 457 </a> 458 459 <footer class="post-card-meta"> 460 <time class="post-card-meta-date" datetime="2025-07-14">14 7ì 2025</time> 461 <span class="post-card-meta-length">16 min read</span> 462
462<script 463 data-ghost-comment-count="6874bc526734f9d66b63dea0" 464 data-ghost-comment-count-empty="" 465 data-ghost-comment-count-singular="comment" 466 data-ghost-comment-count-plural="comments" 467 data-ghost-comment-count-tag="span" 468 data-ghost-comment-count-class-name="" 469 data-ghost-comment-count-autowrap="true" 470> 471</script>
471 472 </footer> 473 474 </div> 475 476</article> 477 </div> 478 </aside> 479 480 481 482 </div> 483 484 <footer class="site-footer outer"> 485 <div class="inner"> 486 <section class="copyright"><a href="https://facerain.me">FacerAin's blog</a> © 2026</section> 487 <div class="site-footer-center"> 488 <div class="site-footer-social-links"> 489 <a href="https://x.com/ghost" target="_blank" rel="noopener" aria-label="X"> 490<svg class="icon" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg"> 491 <path d="M18.2439 2.25H21.5519L14.3249 10.51L22.8269 21.75H16.1699L10.9559 14.933L4.98991 21.75H1.67991L9.40991 12.915L1.25391 2.25H8.07991L12.7929 8.481L18.2439 2.25ZM17.0829 19.77H18.9159L7.08391 4.126H5.11691L17.0829 19.77Z" fill="currentColor"/> 492</svg> </a> 493 <a href="https://www.facebook.com/ghost" target="_blank" rel="noopener" aria-label="Facebook"> 494<svg class="icon" viewBox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" fill="currentColor"><path d="M23.9981 11.9991C23.9981 5.37216 18.626 0 11.9991 0C5.37216 0 0 5.37216 0 11.9991C0 17.9882 4.38789 22.9522 10.1242 23.8524V15.4676H7.07758V11.9991H10.1242V9.35553C10.1242 6.34826 11.9156 4.68714 14.6564 4.68714C15.9692 4.68714 17.3424 4.92149 17.3424 4.92149V7.87439H15.8294C14.3388 7.87439 13.8739 8.79933 13.8739 9.74824V11.9991H17.2018L16.6698 15.4676H13.8739V23.8524C19.6103 22.9522 23.9981 17.9882 23.9981 11.9991Z"/></svg> </a> 495 </div> 496 <nav class="site-footer-nav"> 497 <ul class="nav"> 498 <li class="nav-sign-up"><a href="#/portal/">Sign up</a></li> 499</ul> 500 501 </nav> 502 </div> 503 <div class="gh-powered-by"><a href="https://ghost.org/" target="_blank" rel="noopener">Powered by Ghost</a></div> 504 </div> 505 </footer> 506 507</div> 508 509 <div class="pswp" tabindex="-1" role="dialog" aria-hidden="true"> 510 <div class="pswp__bg"></div> 511 512 <div class="pswp__scroll-wrap"> 513 <div class="pswp__container"> 514 <div class="pswp__item"></div> 515 <div class="pswp__item"></div> 516 <div class="pswp__item"></div> 517 </div> 518 519 <div class="pswp__ui pswp__ui--hidden"> 520 <div class="pswp__top-bar"> 521 <div class="pswp__counter"></div> 522 523 <button class="pswp__button pswp__button--close" title="Close (Esc)"></button> 524 <button class="pswp__button pswp__button--share" title="Share"></button> 525 <button class="pswp__button pswp__button--fs" title="Toggle fullscreen"></button> 526 <button class="pswp__button pswp__button--zoom" title="Zoom in/out"></button> 527 528 <div class="pswp__preloader"> 529 <div class="pswp__preloader__icn"> 530 <div class="pswp__preloader__cut"> 531 <div class="pswp__preloader__donut"></div> 532 </div> 533 </div> 534 </div> 535 </div> 536 537 <div class="pswp__share-modal pswp__share-modal--hidden pswp__single-tap"> 538 <div class="pswp__share-tooltip"></div> 539 </div> 540 541 <button class="pswp__button pswp__button--arrow--left" title="Previous (arrow left)"></button> 542 <button class="pswp__button pswp__button--arrow--right" title="Next (arrow right)"></button> 543 544 <div class="pswp__caption"> 545 <div class="pswp__caption__center"></div> 546 </div> 547 </div> 548 </div> 549</div>
550<script src="/assets/built/casper.js?v=4f24c6eeff" defer></script>
550 551 552ÑÑÑ 553 554</body> 555</html>
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.