TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Implementing Weight-Decomposed Low-Rank Adaptation (DoRA) from Scratch

96 点作者 rasbt超过 1 年前

4 条评论

sorenjan超过 1 年前
Speaking of LoRA, what happened with ZipLoRA? It&#x27;s supposed to be a better way of merging multiple LoRAs, and the results look good in their examples. Is it being used anywhere?<p><a href="https:&#x2F;&#x2F;ziplora.github.io&#x2F;" rel="nofollow">https:&#x2F;&#x2F;ziplora.github.io&#x2F;</a>
评论 #39422792 未加载
gliched_robot超过 1 年前
This is very cool and will change the way we do lora now.
jasonjmcghee超过 1 年前
This is a bit of a misleading title. Why not use the original?<p>&quot;Improving LoRA: Implementing Weight-Decomposed Low-Rank Adaptation (DoRA) from Scratch&quot;<p>(If it&#x27;s too long, just drop the &quot;Improving LoRA: &quot; part)
评论 #39422119 未加载
murkt超过 1 年前
Hooray, no more confusion with LoRa the radio!
评论 #39424091 未加载
评论 #39424277 未加载