ViK

Game Developer + Vehicle Designer | 《Forza》系列开发人员 豪车美术师 退役码农 空间视觉系 爱猫人士

[BuBBliK @k1rallik]

> be Forza Horizon
> starts as a spin-off nobody asked for
> new country every 2-3 years
> FH3, FH4, FH5 - 91, 92, 92 Metacritic in a row
> FH5 - 10 million players week one
> biggest Xbox launch ever. a racing game.
> Japan requested since 2012. 14 years of silence
> Microsoft accidentally leaks the full game on Steam
> modder streams it without blurring his name
> banned until December 31, 9999
> launch day - 273,000 Steam players
> record for any racing game in history
> 6 million players in 72 hours

Forza went from unwanted spin-off to the most consistent franchise in gaming and nobody talks about it.

[Nassim Nicholas Taleb @nntaleb]

More on H1-B:
I've addressed this 18 y ago in The Black Swan: America's edge is in risk taking, not education. Any idiot can have education; you need to learn to fail.
The American system has been based on trial and error. Which is why foreigners succeed here not at home.

[Sabine Hossenfelder @skdh]

I've been telling physicists in the foundations for more than 10 years they need to get their act together or their future looks bleak.

There have been too many unobserved particles, too many idiotic headlines about multiverses, too many expensive experiments that didn't find anything interesting, too many "predictions" that never came to pass -- turning the word "prediction" into a joke -- too much silly talk about beautiful and elegant ideas and theories of everything that remained wishful thinking. And never once did they offer an explanation for what went wrong or how they are going to improve.

Simply put, they have spent too much of taxpayers' money without producing anything of value.

Can you not see the writing on the wall?

Americans are going to crack down hard on academia because of all the nonsense studies that have befallen their universities. They're not going to stop there.

And Europeans are sick and tired of paying ridiculously high amounts of taxes. In the next years you will see a dramatic political shift to deregulation over here too.

In Europe it's hard to fire people but if they get no further funding, many research areas are simply going to die off.

It's not that I wanted this to happen. But this is what I think will happen. And at this point it's too late to do anything about it.

[Rupert Myers @RupertMyers]

The German sense of humour is woefully underrated

[Massimo @Rainmaker1973]

A brief history of physics

― Zach Weinersmith, 'Science: Abridged Beyond the Point of Usefulness'

[Massimo @Rainmaker1973]

The scale of our Solar System vs a section of the Carina Nebula NGC 3324 from one of the first image taken by the James Webb Space Telescope.

Space is big.

[Dulwich Quantum Computing @DulwichQuantum]

"Can Quantum Computers Do Nothing?"
https://t.co/VVF8nopz0M

[ViK @VisualKnight]

@recatm 不能

[John Romero 🤘🏽 @romero]

Finally.

[Jim Fan @DrJimFan]

My TED talk is finally live!! I proposed the recipe for the "Foundation Agent": a single model that learns how to act in different worlds. LLM scales across lots and lots of texts. Foundation Agent scales across lots and lots of realities. If it is able to master 10,000 diverse simulated realities, it may well generalize to our physical world, which is simply the 10,001st reality.

TED talks do not have teleprompters!! All I have is a "confidence monitor" at my foot, showing only the current slide and timer. That means I need to memorize the whole speech. It sounds intimidating at first, but turns out to be the best way to connect with the audience and deliver the ideas right to heart.

The video is only 10 min. I promise it's well worth your time!
https://t.co/sJJdUg2QNQ

Also sharing my slides with all of you! https://t.co/XFvi25fldK

[John Romero 🤘🏽 @romero]

In addition to SIGIL II (https://t.co/Etat3Xllsg), I've got some other great news to celebrate DOOM's 30th Anniversary. Join me and John Carmack @ID_AA_Carmack to discuss DOOM live, moderated by David L. Craddock @davidlcraddock. Dec 10 8pm GMT on https://t.co/r4PKfgG4af. Thanks for playing our games!

[ViK @VisualKnight]

Testing GPT-4V for its vision comprehension. In this particular example I was deeply and terrifyingly impressed!

[Laura Wendel @Lauramaywendel]

How the twitter timeline feels recently

[Programmer Humor @PR0GRAMMERHUM0R]

It wasn't mine in the first place https://t.co/yPO5sNVuhU

[ViK @VisualKnight]

@fi56622380 受Sparks of AGI论文启发，我觉得要提高可靠性不能只依赖算力。那篇论文里提到的“慢思考”模式（借鉴心理学界赫赫有名Daniel Kahneman的研究）才是最有可能大幅降低错误率的方向。当然这是个很���把握的预测，估计需要全新架构被发明出来。

[fin @fi56622380]

更糟糕的是，随着LLM的使用，以后系统监控的理解能力将会到达不可思议的高度

我把这段新闻放进GPT里，GPT能非常轻松的解读出“口罩”背后的含义和背景

从此以后，监控的精细化智能化将会超乎想象，各种别名，代称，表情，符号之类的暗语都逃不过监控了

这意味着以后连“口罩”都将发不出来了

不寒而栗

再细想一下，系统LLM甚至不需要知道你指代了什么，只需要知道你是在表达��满意，不管你隐藏的如何好，都绝无可能逃过LLM的解读

阴阳怪气也好，讽刺也好，调侃也好，河森堡说的“中文社区量子态"也好，顶多稍微加上一点背景，你的立场将会无所遁形，不满意度甚至能给你打出一个精确到小数点后两位的数字

一群人的交流中透露出任何不满意，甚至一群人语言交流中全是”好“，”威武“，”有希望了“这样满意的词语，都将逃不过系统的监控精确的探知你真实的意愿，全部将转换成精确的系统威胁度数据

在1Q84小说中，监控主要还是人力来处理，“小人子”拥有神秘的力量，能进入人的意识。LLM以一种更直观和成本降低了几个数量级的方式，在精细程度上做到了1Q84作者都想象不到的程度

这种事情的发生，在技术上已经是完全摆在我们眼前的事情，根本不是遥远的未来，我们甚至还没有心理准备，这就是最可怕的地方

我们只能祈祷，海归大牛们/国内博士们掌握了LLM技术但没有商业模式的时候，心里还有一点点善念，不要像当年四小龙一样为了营收拼了命的做政府监控项目，给技术以文明，给所有人多留一点活路

PS：量子态的意思是，同样一篇文章，有人读起来觉得是在讽刺，有人读起来就认为是在赞扬，就好像这篇文章的核心论点处于量子叠加态，作者的立场是什么取决于你结合自身生活经历的主观解读

[ViK @VisualKnight]

@fi56622380 很有意思的观察。那么如果未来这些模型能引入某种负反馈，来“反思”自己过去的结论(reflection)，甚至从多随机角度思考起始问题，是否就能控制住指数性错误递增呢��

[ViK @VisualKnight]

In the past few months I’ve been watching this generative AI revolution in absolute astonishment!!

[ViK @VisualKnight]

I care about what future historians will say about the current social movements of our time, and I stand with Thomas Sowell on the matter of Equity. https://t.co/fMvVbjC8Oy