<?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>파운데이션모델 on TechLLM</title><link>https://techllm.github.io/tags/%ED%8C%8C%EC%9A%B4%EB%8D%B0%EC%9D%B4%EC%85%98%EB%AA%A8%EB%8D%B8/</link><description>Recent content in 파운데이션모델 on TechLLM</description><generator>Hugo</generator><language>ko</language><lastBuildDate>Sat, 16 May 2026 10:49:00 +0900</lastBuildDate><atom:link href="https://techllm.github.io/tags/%ED%8C%8C%EC%9A%B4%EB%8D%B0%EC%9D%B4%EC%85%98%EB%AA%A8%EB%8D%B8/index.xml" rel="self" type="application/rss+xml"/><item><title>로봇 AI를 하나로 — Pelican-Unified 1.0, 이해·추론·상상·행동을 단일 모델에</title><link>https://techllm.github.io/posts/pelican-unified-embodied-ai/</link><pubDate>Sat, 16 May 2026 10:49:00 +0900</pubDate><guid>https://techllm.github.io/posts/pelican-unified-embodied-ai/</guid><description>이해, 추론, 상상, 행동을 하나의 체크포인트로 통합한 구현체 AI 파운데이션 모델 Pelican-Unified 1.0. 분리된 전문가 시스템 없이도 8개 VLM 벤치마크 64.7, RoboTwin 93.5, WorldArena 1위를 달성했다.</description></item></channel></rss>