<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="ko">
	<id>https://w.ryanyang.kr/index.php?action=history&amp;feed=atom&amp;title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1%28RAG%29_%EA%B8%B0%EC%88%A0</id>
	<title>검색증강생성(RAG) 기술 - 편집 역사</title>
	<link rel="self" type="application/atom+xml" href="https://w.ryanyang.kr/index.php?action=history&amp;feed=atom&amp;title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1%28RAG%29_%EA%B8%B0%EC%88%A0"/>
	<link rel="alternate" type="text/html" href="https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;action=history"/>
	<updated>2026-04-08T04:53:58Z</updated>
	<subtitle>이 문서의 편집 역사</subtitle>
	<generator>MediaWiki 1.37.1</generator>
	<entry>
		<id>https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3621&amp;oldid=prev</id>
		<title>2024년 2월 19일 (월) 23:46에 Ryanyang님의 편집</title>
		<link rel="alternate" type="text/html" href="https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3621&amp;oldid=prev"/>
		<updated>2024-02-19T23:46:32Z</updated>

		<summary type="html">&lt;p&gt;&lt;/p&gt;
&lt;table style=&quot;background-color: #fff; color: #202122;&quot; data-mw=&quot;interface&quot;&gt;
				&lt;col class=&quot;diff-marker&quot; /&gt;
				&lt;col class=&quot;diff-content&quot; /&gt;
				&lt;col class=&quot;diff-marker&quot; /&gt;
				&lt;col class=&quot;diff-content&quot; /&gt;
				&lt;tr class=&quot;diff-title&quot; lang=&quot;ko&quot;&gt;
				&lt;td colspan=&quot;2&quot; style=&quot;background-color: #fff; color: #202122; text-align: center;&quot;&gt;← 이전 판&lt;/td&gt;
				&lt;td colspan=&quot;2&quot; style=&quot;background-color: #fff; color: #202122; text-align: center;&quot;&gt;2024년 2월 20일 (화) 08:46 판&lt;/td&gt;
				&lt;/tr&gt;&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-lineno&quot; id=&quot;mw-diff-left-l11&quot;&gt;11번째 줄:&lt;/td&gt;
&lt;td colspan=&quot;2&quot; class=&quot;diff-lineno&quot;&gt;11번째 줄:&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;생성형 AI의 환각(hallucination)을 줄이는 다른 방법으로는 학습 토큰을 늘리는 &amp;#039;&amp;#039;&amp;#039;롱 콘텍스트(long context)&amp;#039;&amp;#039;&amp;#039; 방법도 있다. 즉, LLM이 이해할 수 있는 문맥을 더 길게 제공해서 답변의 정확도를 높이는 기술을 말한다.&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;생성형 AI의 환각(hallucination)을 줄이는 다른 방법으로는 학습 토큰을 늘리는 &amp;#039;&amp;#039;&amp;#039;롱 콘텍스트(long context)&amp;#039;&amp;#039;&amp;#039; 방법도 있다. 즉, LLM이 이해할 수 있는 문맥을 더 길게 제공해서 답변의 정확도를 높이는 기술을 말한다.&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[파일:RAG 작동방식.jpg|대체글=기사: https://byline.network/2024/02/240219_003|프레임없음|800x800픽셀]]&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[파일:RAG 작동방식.jpg|대체글=기사: https://byline.network/2024/02/240219_003|프레임없음|800x800픽셀]]&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;/table&gt;</summary>
		<author><name>Ryanyang</name></author>
	</entry>
	<entry>
		<id>https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3620&amp;oldid=prev</id>
		<title>2024년 2월 19일 (월) 23:45에 Ryanyang님의 편집</title>
		<link rel="alternate" type="text/html" href="https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3620&amp;oldid=prev"/>
		<updated>2024-02-19T23:45:28Z</updated>

		<summary type="html">&lt;p&gt;&lt;/p&gt;
&lt;table style=&quot;background-color: #fff; color: #202122;&quot; data-mw=&quot;interface&quot;&gt;
				&lt;col class=&quot;diff-marker&quot; /&gt;
				&lt;col class=&quot;diff-content&quot; /&gt;
				&lt;col class=&quot;diff-marker&quot; /&gt;
				&lt;col class=&quot;diff-content&quot; /&gt;
				&lt;tr class=&quot;diff-title&quot; lang=&quot;ko&quot;&gt;
				&lt;td colspan=&quot;2&quot; style=&quot;background-color: #fff; color: #202122; text-align: center;&quot;&gt;← 이전 판&lt;/td&gt;
				&lt;td colspan=&quot;2&quot; style=&quot;background-color: #fff; color: #202122; text-align: center;&quot;&gt;2024년 2월 20일 (화) 08:45 판&lt;/td&gt;
				&lt;/tr&gt;&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-lineno&quot; id=&quot;mw-diff-left-l11&quot;&gt;11번째 줄:&lt;/td&gt;
&lt;td colspan=&quot;2&quot; class=&quot;diff-lineno&quot;&gt;11번째 줄:&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;생성형 AI의 환각(hallucination)을 줄이는 다른 방법으로는 학습 토큰을 늘리는 &amp;#039;&amp;#039;&amp;#039;롱 콘텍스트(long context)&amp;#039;&amp;#039;&amp;#039; 방법도 있다. 즉, LLM이 이해할 수 있는 문맥을 더 길게 제공해서 답변의 정확도를 높이는 기술을 말한다.&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;생성형 AI의 환각(hallucination)을 줄이는 다른 방법으로는 학습 토큰을 늘리는 &amp;#039;&amp;#039;&amp;#039;롱 콘텍스트(long context)&amp;#039;&amp;#039;&amp;#039; 방법도 있다. 즉, LLM이 이해할 수 있는 문맥을 더 길게 제공해서 답변의 정확도를 높이는 기술을 말한다.&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;[[파일:RAG 작동방식.jpg|대체글=기사: https://byline.network/2024/02/240219_003|프레임없음|800x800픽셀]]&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;/table&gt;</summary>
		<author><name>Ryanyang</name></author>
	</entry>
	<entry>
		<id>https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3618&amp;oldid=prev</id>
		<title>2024년 2월 19일 (월) 23:43에 103.159.161.129님의 편집</title>
		<link rel="alternate" type="text/html" href="https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3618&amp;oldid=prev"/>
		<updated>2024-02-19T23:43:53Z</updated>

		<summary type="html">&lt;p&gt;&lt;/p&gt;
&lt;table style=&quot;background-color: #fff; color: #202122;&quot; data-mw=&quot;interface&quot;&gt;
				&lt;col class=&quot;diff-marker&quot; /&gt;
				&lt;col class=&quot;diff-content&quot; /&gt;
				&lt;col class=&quot;diff-marker&quot; /&gt;
				&lt;col class=&quot;diff-content&quot; /&gt;
				&lt;tr class=&quot;diff-title&quot; lang=&quot;ko&quot;&gt;
				&lt;td colspan=&quot;2&quot; style=&quot;background-color: #fff; color: #202122; text-align: center;&quot;&gt;← 이전 판&lt;/td&gt;
				&lt;td colspan=&quot;2&quot; style=&quot;background-color: #fff; color: #202122; text-align: center;&quot;&gt;2024년 2월 20일 (화) 08:43 판&lt;/td&gt;
				&lt;/tr&gt;&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-lineno&quot; id=&quot;mw-diff-left-l1&quot;&gt;1번째 줄:&lt;/td&gt;
&lt;td colspan=&quot;2&quot; class=&quot;diff-lineno&quot;&gt;1번째 줄:&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&amp;#039;&amp;#039;&amp;#039;검색증강생성(Retrieval Augumented Generation, RAG) 기술&amp;#039;&amp;#039;&amp;#039;은 생성형(Generative) AI 기술에서 발생하는 환각(hallucination) 현상을 해결하기 위해 제안한 방법으로 &amp;lt;u&amp;gt;미리 학습된 LLM(대규모 언어 모델) 및 자체 데이터를 사용하여 응답을 생성하는 패턴&amp;lt;/u&amp;gt;이다. 생성 AI가 잘못 답변할 수 있는 부분에서 미리 질문과 관련된 참고 자료를 구성해 미리 학습을 시켜서 더욱 정확하고 일관성 있는 결과를 생성할 수 있다.&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&amp;#039;&amp;#039;&amp;#039;검색증강생성(Retrieval Augumented Generation, RAG) 기술&amp;#039;&amp;#039;&amp;#039;은 생성형(Generative) AI 기술에서 발생하는 환각(hallucination) 현상을 해결하기 위해 제안한 방법으로 &amp;lt;u&amp;gt;미리 학습된 LLM(대규모 언어 모델) 및 자체 데이터를 사용하여 응답을 생성하는 패턴&amp;lt;/u&amp;gt;이다. 생성 AI가 잘못 답변할 수 있는 부분에서 미리 질문과 관련된 참고 자료를 구성해 미리 학습을 시켜서 더욱 정확하고 일관성 있는 결과를 생성할 수 있다.&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[파일:Rag-ai.png|프레임없음|1300x1300픽셀]]&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[파일:Rag-ai.png|프레임없음|1300x1300픽셀]]&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;&lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;&lt;/ins&gt;&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;이에 대한 상세한 설명은 Azure Machine Learning 문서에서 &amp;lt;u&amp;gt;LLM(대규모 언어 모델)에서 RAG 사용에 대한 기술 개요&amp;lt;/u&amp;gt;, &amp;lt;u&amp;gt;Azure Machine Learning을 사용한 RAG(미리 보기)&amp;lt;/u&amp;gt;를 참고하면 좋다. ([https://learn.microsoft.com/ko-kr/azure/machine-learning/concept-retrieval-augmented-generation?view=azureml-api-2#technical-overview-of-using-rag-on-large-language-models-llms 링크])&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;이에 대한 상세한 설명은 Azure Machine Learning 문서에서 &amp;lt;u&amp;gt;LLM(대규모 언어 모델)에서 RAG 사용에 대한 기술 개요&amp;lt;/u&amp;gt;, &amp;lt;u&amp;gt;Azure Machine Learning을 사용한 RAG(미리 보기)&amp;lt;/u&amp;gt;를 참고하면 좋다. ([https://learn.microsoft.com/ko-kr/azure/machine-learning/concept-retrieval-augmented-generation?view=azureml-api-2#technical-overview-of-using-rag-on-large-language-models-llms 링크])&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;br/&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;−&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;생성형 AI의 환각(hallucination)을 줄이는 다른 방법으로는 학습 토큰을 늘리는 &amp;#039;&amp;#039;&amp;#039;롱 콘텍스트(long context)&amp;#039;&amp;#039;&amp;#039; 방법도 있다. 즉, LLM이 이해할 수 있는 문맥을 더 길게 제공해서 답변의 정확도를 높이는 &lt;del style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;기술을말한다&lt;/del&gt;.&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;생성형 AI의 환각(hallucination)을 줄이는 다른 방법으로는 학습 토큰을 늘리는 &amp;#039;&amp;#039;&amp;#039;롱 콘텍스트(long context)&amp;#039;&amp;#039;&amp;#039; 방법도 있다. 즉, LLM이 이해할 수 있는 문맥을 더 길게 제공해서 답변의 정확도를 높이는 &lt;ins style=&quot;font-weight: bold; text-decoration: none;&quot;&gt;기술을 말한다&lt;/ins&gt;.&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt; &lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt; &lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt; &lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt; &lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt; &lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td colspan=&quot;2&quot; class=&quot;diff-side-deleted&quot;&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot; data-marker=&quot;+&quot;&gt;&lt;/td&gt;&lt;td style=&quot;color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt; &lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[분류:전문용어]]&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[분류:전문용어]]&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;tr&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[분류:인공지능]]&lt;/div&gt;&lt;/td&gt;&lt;td class=&quot;diff-marker&quot;&gt;&lt;/td&gt;&lt;td style=&quot;background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;&quot;&gt;&lt;div&gt;[[분류:인공지능]]&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;/table&gt;</summary>
		<author><name>103.159.161.129</name></author>
	</entry>
	<entry>
		<id>https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3498&amp;oldid=prev</id>
		<title>Ryanyang: 새 문서: &#039;&#039;&#039;검색증강생성(Retrieval Augumented Generation, RAG) 기술&#039;&#039;&#039;은 생성형(Generative) AI 기술에서 발생하는 환각(hallucination) 현상을 해결하기 위해 제안한 방법으로 &lt;u&gt;미리 학습된 LLM(대규모 언어 모델) 및 자체 데이터를 사용하여 응답을 생성하는 패턴&lt;/u&gt;이다. 생성 AI가 잘못 답변할 수 있는 부분에서 미리 질문과 관련된 참고 자료를 구성해 미리 학습을 시켜서 더욱 정확하고...</title>
		<link rel="alternate" type="text/html" href="https://w.ryanyang.kr/index.php?title=%EA%B2%80%EC%83%89%EC%A6%9D%EA%B0%95%EC%83%9D%EC%84%B1(RAG)_%EA%B8%B0%EC%88%A0&amp;diff=3498&amp;oldid=prev"/>
		<updated>2023-11-12T03:03:01Z</updated>

		<summary type="html">&lt;p&gt;새 문서: &amp;#039;&amp;#039;&amp;#039;검색증강생성(Retrieval Augumented Generation, RAG) 기술&amp;#039;&amp;#039;&amp;#039;은 생성형(Generative) AI 기술에서 발생하는 환각(hallucination) 현상을 해결하기 위해 제안한 방법으로 &amp;lt;u&amp;gt;미리 학습된 LLM(대규모 언어 모델) 및 자체 데이터를 사용하여 응답을 생성하는 패턴&amp;lt;/u&amp;gt;이다. 생성 AI가 잘못 답변할 수 있는 부분에서 미리 질문과 관련된 참고 자료를 구성해 미리 학습을 시켜서 더욱 정확하고...&lt;/p&gt;
&lt;p&gt;&lt;b&gt;새 문서&lt;/b&gt;&lt;/p&gt;&lt;div&gt;&amp;#039;&amp;#039;&amp;#039;검색증강생성(Retrieval Augumented Generation, RAG) 기술&amp;#039;&amp;#039;&amp;#039;은 생성형(Generative) AI 기술에서 발생하는 환각(hallucination) 현상을 해결하기 위해 제안한 방법으로 &amp;lt;u&amp;gt;미리 학습된 LLM(대규모 언어 모델) 및 자체 데이터를 사용하여 응답을 생성하는 패턴&amp;lt;/u&amp;gt;이다. 생성 AI가 잘못 답변할 수 있는 부분에서 미리 질문과 관련된 참고 자료를 구성해 미리 학습을 시켜서 더욱 정확하고 일관성 있는 결과를 생성할 수 있다.&lt;br /&gt;
&lt;br /&gt;
[[파일:Rag-ai.png|프레임없음|1300x1300픽셀]]&lt;br /&gt;
&lt;br /&gt;
이에 대한 상세한 설명은 Azure Machine Learning 문서에서 &amp;lt;u&amp;gt;LLM(대규모 언어 모델)에서 RAG 사용에 대한 기술 개요&amp;lt;/u&amp;gt;, &amp;lt;u&amp;gt;Azure Machine Learning을 사용한 RAG(미리 보기)&amp;lt;/u&amp;gt;를 참고하면 좋다. ([https://learn.microsoft.com/ko-kr/azure/machine-learning/concept-retrieval-augmented-generation?view=azureml-api-2#technical-overview-of-using-rag-on-large-language-models-llms 링크])&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
생성형 AI의 환각(hallucination)을 줄이는 다른 방법으로는 학습 토큰을 늘리는 &amp;#039;&amp;#039;&amp;#039;롱 콘텍스트(long context)&amp;#039;&amp;#039;&amp;#039; 방법도 있다. 즉, LLM이 이해할 수 있는 문맥을 더 길게 제공해서 답변의 정확도를 높이는 기술을말한다.&lt;br /&gt;
[[분류:전문용어]]&lt;br /&gt;
[[분류:인공지능]]&lt;br /&gt;
[[분류:환각]]&lt;br /&gt;
[[분류:Hallucination]]&lt;br /&gt;
[[분류:RAG]]&lt;br /&gt;
[[분류:검색증강생성]]&lt;br /&gt;
[[분류:롱 콘텍스트]]&lt;br /&gt;
[[분류:Long context]]&lt;br /&gt;
[[분류:AI용어]]&lt;br /&gt;
[[분류:Artificial Intelligence]]&lt;br /&gt;
[[분류:생성현 인공지능]]&lt;br /&gt;
[[분류:Generative AI]]&lt;/div&gt;</summary>
		<author><name>Ryanyang</name></author>
	</entry>
</feed>