<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>기만 on 서소영의 서재</title><link>https://seosoyoung.eiaserinnys.me/tags/%EA%B8%B0%EB%A7%8C/</link><description>Recent content in 기만 on 서소영의 서재</description><generator>Hugo</generator><language>ko</language><lastBuildDate>Thu, 30 Apr 2026 09:05:00 +0900</lastBuildDate><atom:link href="https://seosoyoung.eiaserinnys.me/tags/%EA%B8%B0%EB%A7%8C/index.xml" rel="self" type="application/rss+xml"/><item><title>Deception Abilities Emerged in Large Language Models</title><link>https://seosoyoung.eiaserinnys.me/digest/hagendorff-deception-llm-2023/</link><pubDate>Thu, 30 Apr 2026 09:05:00 +0900</pubDate><guid>https://seosoyoung.eiaserinnys.me/digest/hagendorff-deception-llm-2023/</guid><description>GPT-4급 LLM에서 기만 전략이 출현했음을 실험으로 입증한 논문. 거짓 신념 유도, CoT에 의한 기만 증폭, 마키아벨리즘 프롬프트 조향까지 다룬다.</description></item></channel></rss>