The opposite state will be called generalized focus.

与之相反的状态叫做泛化专注。

Humans are a generalist species because we eat almost anything.

人类是一个泛化种，因为我们几乎什么都吃。

And this allows it to generalize.

这样 AI 就可以泛化。

Generalized Focus is broadly distributed attention.

泛化专注即宽广分散的注意力。

And so you can see that it's really learning the process, but it hasn't fully generalized, right?

你可以看出，它在学习这个过程，但还没有完全泛化，对吧？

" Learning from less variable input is often fast, but may fail to generalise to new stimuli."

“从较少变的输入中学习通常很快，但可能无法泛化为新的刺激。”

It will enter into a state of generalized focus, determine an imminent threat, and begin saving people or evacuating.

这让机器人进入泛化专注状态，近威胁并开始救人或撤离。

For example, generalists may be able to tolerate a wide range of temperatures or have a varied diet.

例如，泛化种可能能够忍受较大的温度范围或具有多样化的饮食。

Generalists cope better because they can adapt when the environment changes.

泛化种能够很好地应对，因为他们能适应环境的变化。

One interesting notion is that of stimulus generalization.

一个有趣的概念是刺激泛化。

Luckily, we've now done 95% of the work, generalizing is very brief.

幸运的是，我们现在已经完成了95%的工作，泛化部分非常简短。

And showed that they didn't generalize essentially beyond the distribution.

并且表明它们基本上无法超越数据分布进行泛化。

Sophie says we're quick to generalize or to give our clothes too much power.

索菲说我们很容易泛化，或者给衣服太多力。

And how they generalized a prediction for half a day to a more specific time.

以及他们如何将半天的预泛化到更具体的时间。

So I'm sure that's an overgeneralization, but that resonates for a lot of us.

所以我知道那是一种泛化，但这一点对很多人来说很有共鸣。

And it was in some sense genericized, if that's a word.

从某种意义上说，它被泛化了，如果这个词合适的话。

And... Which any generalization is technically wrong.

而且… … 任何泛化从技术上讲都是错误的。

One of the things about the G and AGI is about generalization.

关于大语言模型和通用人工智能的一个特点就是泛化能力。

But at the fundamental level, what we're talking about are these overgeneralized beliefs.

但在根本层面，我们讨论的是这些过度泛化的信念。

There's a lot of overlap, but it's a classic ML topic around generalization.

虽然有很多重叠之处，但这实际上是一个关于泛化的经典机器学习课题。