<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Yuyang Dong — 博客</title><link>https://dongyuyang.github.io/zh/blog/</link><description>Yuyang Dong (David Dong) — 日语基础模型、数据中心 AI、多模态理解与智能体研究。</description><language>zh</language><atom:link href="https://dongyuyang.github.io/zh/feed.xml" rel="self" type="application/rss+xml"/><item><title>AI 越来越好用了，然后呢？</title><link>https://dongyuyang.github.io/zh/blog/ai-keeps-getting-better/</link><guid>https://dongyuyang.github.io/zh/blog/ai-keeps-getting-better/</guid><description>做模型、用 AI，也有点焦虑。关于年轻人、被磨平的门槛，以及睡前看的宇宙小知识。</description><pubDate>Mon, 12 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Sarashina3 mini / nano</title><link>https://www.sbintuitions.co.jp/blog/entry/2026/06/30/100000</link><guid>https://www.sbintuitions.co.jp/blog/entry/2026/06/30/100000</guid><description>预训练、强化学习与智能体能力。发表于 SB Intuitions 官方技术博客。</description><pubDate>Tue, 30 Jun 2026 00:00:00 GMT</pubDate></item><item><title>大语言模型在表格数据处理中的应用</title><link>https://dongyuyang.github.io/papers/tablellm_%E6%9C%80%E5%BC%B7DB_YuyangDong.pdf</link><guid>https://dongyuyang.github.io/papers/tablellm_%E6%9C%80%E5%BC%B7DB_YuyangDong.pdf</guid><description>最强 DB 讲座第 37 期的演讲资料。</description><pubDate>Wed, 09 Jul 2025 00:00:00 GMT</pubDate></item><item><title>多模态模型融合尝试：VLM 与 LLM 之间的选择性参数融合</title><link>https://zenn.dev/dongyuyang/articles/de0f432de66dd6</link><guid>https://zenn.dev/dongyuyang/articles/de0f432de66dd6</guid><description>Qwen2.5-VL 与 Qwen2.5 的模型融合实验，比较仅合并 self-attention 参数及分层权重策略。发表于 Zenn，原文为日语。</description><pubDate>Wed, 02 Apr 2025 00:00:00 GMT</pubDate></item><item><title>NEC 的日语大模型开发</title><link>https://www.nec.com/en/global/rd/technologies/202310/index.html</link><guid>https://www.nec.com/en/global/rd/technologies/202310/index.html</guid><description>开发团队介绍训练数据、模型设计与评估。</description><pubDate>Fri, 01 Sep 2023 00:00:00 GMT</pubDate></item></channel></rss>
