TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

OpenAI funded independent math benchmark before setting record with o3

56 pointsby rar004 months ago

5 comments

andrepd4 months ago
&gt; They also made a verbal agreement with OpenAI that prohibits the company from using the materials to train their models<p>Hilarious.
aithrowawaycomm4 months ago
Elliot Glazer seems to have been caught in a contradiction: <a href="https:&#x2F;&#x2F;xcancel.com&#x2F;ElliotGlazer&#x2F;status&#x2F;1880809468616950187" rel="nofollow">https:&#x2F;&#x2F;xcancel.com&#x2F;ElliotGlazer&#x2F;status&#x2F;1880809468616950187</a><p>Here he says that Epoch is &quot;developing&quot; a private test set that OpenAI doesn&#x27;t have access to, but elsewhere Epoch strongly implied that this already existed. This kind of makes me lean towards &quot;Epoch AI lied&quot; instead of &quot;Epoch AI got played.&quot; (Even the coauthors weren&#x27;t informed about the funding, so Epoch does not deserve a presumption of good faith.)<p>I guess the real question: o3 was able to solve 25% of Frontier problems, so were these the problems whose solutions OpenAI had access to? If so, then that score is meaningless and dishonest.
ChrisArchitect4 months ago
[dupe] Discussion on source:<p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=42763231">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=42763231</a>
nioj4 months ago
Related <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=42763231">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=42763231</a>
Frederation4 months ago
Cant trust anyone, ever.
评论 #42771554 未加载