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。

Full draft of Timnit Gebru's paper

25 点作者 organian超过 4 年前

3 条评论

gedy超过 4 年前
The issue I take with the concerns raised in the paper regarding:<p>&quot;A language model that has been trained on such data will pick up these kinds of problematic associations.&quot;<p>&quot;Such data&quot; is basically everyday, normal discourse, and some of the &quot;problematic associations&quot; are training that includes phrases like &quot;woman doctor&quot;, &quot;both genders&quot;, etc. While I get the point, this itself is a biased interpretation of discourse and would be worrisome imho to have people filtering models with their own biases vs the language as it&#x27;s used by the vast majority of people.
评论 #25326290 未加载
dredmorbius超过 4 年前
Link disintermediation: <a href="https:&#x2F;&#x2F;gofile.io&#x2F;d&#x2F;WfcxoF" rel="nofollow">https:&#x2F;&#x2F;gofile.io&#x2F;d&#x2F;WfcxoF</a>
moomin超过 4 年前
Obviously, bear in mind this is a(n early) draft. That means that _it’s not finished_. You really wouldn’t want to read the first cut of Lord of the Rings (yeah, I know some of you would).
评论 #25326796 未加载
评论 #25326827 未加载