Formasan Frog

[Formasan Frog @FormosanFrog]

@youchaowoo @Kay2289123 @grok 为什么openai官方说用了Cerebras（https://t.co/XlHH2c2FXI）这个人却说没有用？

[ali(gnment) @waterloo_intern]

https://t.co/aivc7B4A7Z

[最爱吃兽奶的兔🐰 @0xMilkRabbit]

卧槽，这届群友执行力太强了…

汇丰 Pulse 卡一周就下卡了！

群友反馈：
7 月 6 日看到🐰的提醒开始搬砖，
7 月 8 日搬够 13w 立马申请 Pulse

没想到今天就通过啦！

虽然是地板额度，
不过后面提额的方法很多，而且还可以存刷，最重要的是先下卡！

再次提醒，
还没申请的小伙伴赶紧申请，
趁现在汇丰还有迎新活动，直接拿 1800rc（和 rmb 约 1:1 兑换）

现在距离汇丰迎新活动结束还有 2 周！

记得注意👇:
1）申请的时候尽量填写无业（与实际情况不冲突）
2）填写收入控制在 2w-2.5w
3）内地消费返现 top 级卡，而且没有年费！
4）申请完成后，填无业也会提醒上传资料，可以直接忽略不影响下卡
5）迎新活动记的填活动码：HSBCFLASH
6）还没审批的卓越账户申请，想拿奖励的可以撤��重新申请，
one 账户不推荐，毕竟还是以下卡为主！

最后，
依旧老生常谈，很多事赶早不赶晚！
没申请的小伙伴抓紧申请哦🫶

[前端哥Liam @jinglian]

https://t.co/lFW785zEfE

[Andrej Karpathy @karpathy]

This works really well btw, at the end of your query ask your LLM to "structure your response as HTML", then view the generated file in your browser. I've also had some success asking the LLM to present its output as slideshows, etc.

More generally, imo audio is the human-preferred input to AIs but vision (images/animations/video) is the preferred output from them. Around a ~third of our brains are a massively parallel processor dedicated to vision, it is the 10-lane superhighway of information into brain. As AI improves, I think we'll see a progression that takes advantage:

1) raw text (hard/effortful to read)
2) markdown (bold, italic, headings, tables, a bit easier on the eyes) <-- current default
3) HTML (still procedural with underlying code, but a lot more flexibility on the graphics, layout, even interactivity) <-- early but forming new good default
...4,5,6,...
n) interactive neural videos/simulations

Imo the extrapolation (though the technology doesn't exist just yet) ends in some kind of interactive videos generated directly by a diffusion neural net. Many open questions as to how exact/procedural "Software 1.0" artifacts (e.g. interactive simulations) may be woven together with neural artifacts (diffusion grids), but generally something in the direction of the recently viral https://t.co/z21CP5iQfu

There are also improvements necessary and pending at the input. Audio nor text nor video alone are not enough, e.g. I feel a need to point/gesture to things on the screen, similar to all the things you would do with a person physically next to you and your computer screen.

TLDR The input/output mind meld between humans and AIs is ongoing and there is a lot of work to do and significant progress to be made, way before jumping all the way into neuralink-esque BCIs and all that. For what's worth exploring at the current stage, hot tip try ask for HTML.

[思维怪怪 @0xLogicrw]

OpenAI 后训练核心成员翁家翌（Jiayi Weng）以个人名义提出了一种名为「启发式学习」的强化学习新范式，并开源了全部实验代码。他用 Codex（GPT-5.4）反复玩 Atari 打砖块游戏，但 GPT-5.4 自始至终没有被重新训练过。真正在进步的，是 GPT-5.4 写出来的那套游戏策略代码。

流程是这样的：GPT-5.4 先写一版打砖块的 Python 策略，跑一局，看录像，找出哪里打丢了球，然后自己改代码再跑。经过几轮迭代，策略代码从 387 分涨到了 864 满分。全程没有任何神经网络被训练，纯靠 AI 反复修改 if-else 规则、调落点预测、加死循环检测。最终那套代码包含球路预测器、卡球检测器、回归测试和实验日志，已经长成了一个完整的软件系统。

这和传统强化学习的核心区别在于「学到的东西存在哪」。传统做法把知识压进神经网络参数里，人看不懂，学新任务还容易把旧的覆盖掉（即灾难性遗忘）。翁家翌的做法反过来：知识就是代码，人能读、能改、能加测试锁住，不会因为学新东西就丢了旧本领。

除了打砖块满分，他还在 MuJoCo Ant（模拟机器蚂蚁走路）上跑出超 6000 分的深度强化学习级成绩，在 Atari57 全套 57 个游戏上逼近了 PPO 基准。但翁家翌也明确画了边界：纯代码搞不定复杂感知任务，比如用 Python 写 if-else 去认图片。

他设想的终局是混合架构：��层用轻量神经网络负责视觉等感知，中层用启发式学习处理实时逻辑和安全规则，顶层由大模型审查日志、改代码，再周期性地用底层积累的高质量数据更新自身。过去手写规则之所以被淘汰，不是因为规则没用，而是人类维护不起。现在 AI 写代码够快够好，这条老路重新走得通了。

[Formasan Frog @FormosanFrog]

@CCutieeePIE @SFumoto 👍多么智慧的见解

[Akshay 🚀 @akshay_pachaar]

https://t.co/SSSIK3BX4z

[Thariq @trq212]

https://t.co/45C3gKydTK

[耳朵 @RookieRicardoR]

我把 Andrej Karpathy 推荐的长文章 RSS 订阅源，整理成了一个 SKILL。

一共有 90 个源，更新也不多���但是质量都是经过 AK 把关的，我又额外添加了评分机制，会把每次新更新进行 1 - 10 打分，只输出 7 分以上的文章地址和推荐语。

安装命令：npx skills add rookie-ricardo/erduo-skills --skill ak-rss-digest

github：https://t.co/83inIUDqQw

[Viv @Vtrivedy10]

https://t.co/Lb2gtM9SNs

[Formasan Frog @FormosanFrog]

@bookwormengr @natolambert @Kimi_Moonshot @grok fact check

[Formasan Frog @FormosanFrog]

@joaasew @vermoutharc @shuaiqiMr 抖音博主 小鱼宝

[Formasan Frog @FormosanFrog]

@SpaceGliderX @askOkara Way better than a human who only knows about one single event.

[Andrej Karpathy @karpathy]

New art project.
Train and inference GPT in 243 lines of pure, dependency-free Python. This is the *full* algorithmic content of what is needed. Everything else is just for efficiency. I cannot simplify this any further.
https://t.co/HmiRrQugnP