AI 与科学发现

事实证明，为了让强化学习在大规模下取得良好效果，你确实需要自己对模型进行预训练。你不得不这么做，这两者之间的耦合太紧密了，你必须两者兼顾。

it turned out you actually do need to pre-train your model in order to make reinforcement learning work very well at scale. So you actually kind of do need to do that the things are just so tightly coupled that uh you you do need to do both.

语境：Misha Laskin 解释为何 ReflectionAI 不能只做微调或后训练：基础预训练与强化学习表征存在深度协同，必须端到端闭环开发。

支持 → dwarkesh-podcast-ajeya-cotra-20260901

当我最初启动这些智能体时，我原本期望它们能构想出某些前所未见、颠覆认知的神奇优化器思路，但实情并非如此。它们擅长玩一些巧妙的花样，将不同论文的方法拼接组合并做出渐进改良，但并没能真正发明出任何全新的优化器。

when I launched this these different agents I was expecting them to come up with some crazy ideas on optimizer that's like no one of discover but honestly it wasn't the case. Uh they did some clever trick where basically they combine different papers. uh they kind of do plus one improvement over a bunch of method but there was really like no novel optimizer

语境：Elie Bakouch 客观界定当前 AI 科学家的能力边界：在已知知识库中执行高密度启发式搜索极其强悍，但在无先验指导下从零发明新范式依然受限。

支持 → ai-engineer-elie-bakouch-20260926

Arya 擅长编排作业、理解 GPU 负载、响应 W&B 生态内的训练事件、倾听研究员需求、检索 arXiv 论文并协作推敲假设。我们可以让 Arya 全权接管你不想处理的繁重执行机械工作，而让你集中精力去探索全新的思想、新架构以及前沿实验参数。

Arya's great at orchestrating jobs, understanding GPU workloads, responding to events within the within the Wandi ecosystem, and listening to researchers, uh, uh, looking up archive papers, and collaborating on hypothesis. So, we can let Arya drive the mechanics that you don't want to deal with while you focus on the new ideas, new architectures, and new parameters that you wanted to try.

语境：Tim Sweeney 阐述 AI 科研伴侣的落地愿景：智能体充当实验室管家与执行双手，将人类科学家从集群配置与调参监工中解放出来。

支持 → all-in-satya-nadella-20260915

这一时期的绝大多数视觉 AI 仅仅是靠像素效果来被评判的。如果你生成了漂亮的像素，你就做出了看似精美的软件，但这些像素仅仅是在尝试模拟观测现象。它们并不实际具备或表征现实世界背后的任何语义。因此，它们根本无法带来真正的真实性，无法作为规划的依据，也无法作为对真实物理世界的稳健仿真基础设施。

most of visual AI for this period has been just judged by the pixels. If you've got beautiful pixels, you've got a beautiful piece of software, but these pixels are trying to simulate observations. They don't actually have or represent any of the semantics behind the world. And therefore, they aren't good at realism. They aren't good for giving a basis to plan. They aren't good for actually having a robust simulation infrastructure of the real world.

语境：Christopher Manning 尖锐批评视觉生成与世界模型混为一谈的现状：像素级的逼真无法替代底层因果规律，缺乏因果表征的模型无法用于智能体规划。

支持 → latent-space-anastasis-germanidis-20260925

支持 → ai-engineer-deepak-pathak-20260924

这种神经符号方法你可以称之为保守路线：在动作的生成与控制中明确借助物理引擎和物理学知识。当你面对一张水的图像并希望理解其运动规律时，你是通过物理模型来预测其行为表现的。

this neurosyolic approach is you know you call a conservative approach if you will you know that this is absolutely using physics engines and knowledge of physics um to in its generation and control of movement, right? That if you're what you know when you've just started off with an image of water or and you want to understand how that's behaves, you're using a physics model to predict how it's going to behave.

语境：Christopher Manning 阐述神经符号系统在物理建模中的必要性：将已有物理知识显式嵌入生成与控制循环，而非寄希望于黑箱网络完全隐式重学物理学规律。

支持 → latent-space-anastasis-germanidis-20260925

从根本上讲，这里存在着一种典型的军备竞赛动态，而防御端此前一直远远落后。我们要做的是让防御能力迎头赶上。作为基础研究实验室，我们坚信唯有掌握顶尖序列生成设计能力的团队，才最具备构建底层防御能力的资质，因为二者在模型底层本质上同出一源。

So I think inherently there is this arms race style dynamic that the defensive side has been far far lagging. And so what we want to do is bring the defensive side [music] to par. Essentially, we felt it was important as a lab that a team that was both building the design capabilities is actually also best suited for building the defense capabilities cuz they're basically the same models.

语境：Eric Nguyen 揭示生成式生物安全的核心悖论：能生成强效功能序列的模型天然具备判别潜在病原体的最佳表征，攻防一体的双重使命是生物模型团队不可推卸的责任。

支持 → all-in-satya-nadella-20260915

TabPFN 是 AutoML 演进的自然终局：我们学习的是整个能在单次前向传播中执行的完整学习算法；根据你输入的具体数据集，它会在上下文内部拟合出完全不同的分类器。它不再是单纯的一个静态分类模型，而是一个能根据输入即时生成不同分类器的超网络。

TPFn is really this natural progression of AutoML where we learn this entire algorithm that is executed in a forward pass and depending on the data set that you feed in it learns a different classifier. So um it's not just one network that is one classifier but it's a a network that can learn different classifiers depending on the input

语境：Frank Hutter 阐述 TabPFN 的本质跃迁：从传统反复试错的模型训练演变为上下文学习算法，将整个机器学习拟合过程转化为 Transformer 的单次前向推理。

支持 → all-in-satya-nadella-20260915

预测模型就像是：你输入一些数据，得到一些输出……那仅仅是一个统计模型。而解释性模型才是科学真正追求的目标——它必须能够进行外推，因为它内部捕获了物理规律或对底层现实的某种真实描述。

a predictive model is like let's say you just have a you have some inputs and you have some outputs […] That's just a statistical model, right? A descriptive model is actually what science is trying to get to, which is okay, it should be able to extrapolate because it has sort of the physics or the actual some description of reality that's captured within it.

语境：Google Fellow John Platt 剖析统计机器学习与真实科学研究的根本分界：拟合历史数据的预测模型只能内插，而科学的核心是发现包含物理因果机制、具备跨领域极端条件外推能力的解释性模型。

支持 → youtube-terence-tao-20260828

限定 → youtube-demis-hassabis-20260911

到目前为止，我们还无法构建出一个能够真正发现全新物理学或全新科学理论的系统，但它确实是一种能够帮助人类科学家发现全新科学的强大电动工具。

We haven't been able to make a system so far […] that can really sort of discover completely new physics or completely new science, but this is sort of a power tool to help you discover completely new science.

语境：对当前‘AI 科学家’神话进行清醒解构：现存架构无法从零归纳出革命性的底层物理定律，其实质是极大加速探索效率的‘动力工具’，真正的因果假设与严格证伪仍完全依赖人类科学家。

支持 → youtube-terence-tao-20260828

限定 → sequoia-sutton-20260818

这些 AI 工具就像直升机，能直接把你空运到指定的瀑布前，让你看一眼然后飞回去，但你对如何抵达那里一无所知。除了你具体要求的那个事物之外，你可能看不到任何其他有趣的现象。因此，尽管技术上你更高效地达成了目标，但也许有些东西已经失落了。

these AI tools, they can be like helicopters that will just fly you directly to this waterfall and you can see it and then you fly back, but you learn nothing about how to get there. You may not see any other interesting phenomena than the specific thing that you asked for. And so even though technically you achieve your goal much more efficiently, there may be something that is lost.

语境：陶哲轩提出著名的‘直升机隐喻’，反思 AI 辅助科学发现的潜在代价：自动化工具虽能高效求解特定目标，但跳过了人类徒步穿越数学丛林中附带获得的整体洞察与知识地貌认知。

回应 → youtube-demis-hassabis-20260911

支持 → the-economist-yuval-harari-20260826

我们现在正在生成大量的证明，也验证了其中不少；但理解这些证明并将其整理成最终教科书形式的节奏，依然由人类掌控。事实上，我们现在正在经历所谓的‘证明消化不良’：突然之间涌现出海量待理解、待写入教材的问题解答，但我们已经被过于庞大的数量彻底淹没了。

We are now generating many proofs. We are verifying a bunch of them but the pace of understanding them and putting them into the final textbook form is still done by humans. In fact, we're now experiencing what you might call proof indigestion where suddenly there's lots and lots of pending solutions to problems that should be understood and should be going to textbooks, but we're just flooded now with too many of them.

语境：陶哲轩指出数学界正在遭遇的知识瓶颈：AI 极大加速了前端的蛮力命题生成与形式化验证，但后端的语义消化、逻辑重构与教材沉淀依然受制于人类认知带宽。

限定 → dwarkesh-podcast-noam-brown-20260917

解决千禧年大奖难题（纳维-斯托克斯方程）的成果，绝不能归功于多智能体系统，我甚至连 10% 的功劳都不会分给多智能体。现实是 OpenAI 训练了一个非常强大的模型。我们能让该模型在超长跨度内运行并并行思考，但其根本原因在于我们拥有一个极其强大的通用基础模型。

The effort to solve a Millennium Prize Problem, this was not due to multi-agent. I wouldn’t even attribute 10% of the credit to multi-agent. The reality is that OpenAI has trained a very powerful model. We can get that model to operate over very long horizons. We can get it to think in parallel. But at its core, the reason why we’re able to do this is because we just have a general-purpose, very strong model.

语境：针对业界过度炒作‘万级智能体协同解题’的热潮，Noam Brown 澄清技术实质：多智能体外壳只是显眼包装，真正实现数学与科学发现质变的是底层通用基础模型的长程推理能力。

回应 → tbpn-sam-altman-20260929

支持 → youtube-demis-hassabis-20260911