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。

Ask HN: Have you noticed decreased quality in GPT-4 reasoning recently?

4 点作者 ammar_x将近 2 年前
I noticed a significant decrease in the model reasoning abilities in the last few days, especially when it comes to writing code. Has anyone else experienced this?<p>Given that these models are being used in all sorts of applications all over the world, I think it&#x27;s concerning that OpenAI can lobotomize a model in secret and no one can verify that and do something about it.

3 条评论

skilled将近 2 年前
As an EU user, I feel like it has been watered down a lot, both 3.5 and 4. Many queries now come with disclaimers which are pretty annoying since you can&#x27;t set your own system instructions in the ChatGPT interface. Using the API is fine though since you can explicitly say that you want a specific format.
评论 #35959110 未加载
rainy59将近 2 年前
The more they force the model into politically-fashionable responses, the further they get away from rational thought process. CodeT5+ looks promising though
mdp2021将近 2 年前
&gt; <i>concerning</i><p>...And in fact some at Google are already projecting a future of Open Source dominance in the subfield.<p>(Edit: the system must have missed a post update: )<p>...Coincidentally, in fact, a neighbouring post is: <i>OpenAI readies new open-source AI model</i> - <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=35958789" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=35958789</a>