舞台魔术师经常用多种方式表演同样的魔术。 一个迭代使用 gi道具，下一个显示您的道具不是 gi道具，当你看表演时，你假设只有一种技术，所以每次他们这样做，他们似乎都会揭穿你关于它必须如何运作的假设。
这几乎就是`Astra'的实验可以读写重量来改进Qwen的后期工作。 这篇文章的抽象称：
GPT-6 Astra通过观察原始的重量，可以知道一个网络是怎样训练的。 而没有任何环境反馈，它没有写作困难，从零开始，小游戏播放网络超越了PPO训练的1亿步的游戏网络。
同样，Astra要求让Qwen3.5-0.8B在数学上好一些，而不是梯度。 Astra在权重中找出一个问题，引起Qwen自己的模型卡警告左右的推理循环，并通过本土编辑0.016%的权重来修正这个问题。 在相同的采样设置下，这种重量校正融资 MATH-500从7.9%升至48.1%，并概括出框到后面的指令，将IFEval提高3x.
Qwen实验“只看原始重量”有效吗？事实并非如此。 事实上，即使是“小游戏网络”实验也“仅看原始重量”行不通。 也有研究原始重量的实验，改进游戏游戏网络的实验，还有在数学上改进Qwen的实验。 但后两者并不只是看原始重量。
以下是该职位的实际工作以及技术如何运作。
" 仅看原始重量 " 设定(上下文中的浮点、没有工具、没有运行网络)只在实验1、2和4中出现在17K-172K重量的网络中。 每个涉及真语言模型的实验都使用Python，而Qwen结果作为头条，也可以在多达50批的练习问题上运行模型。
Qwen的结果其实是显示可以促使Astra设计和完成人类ML工程师可以完成的任务。 人类ML工程师编写代码片段，这些代码分析甚至修改网络权重，"直接"不断地。 找出如何卓有成效地做到这一点并不是什么事情，一年或两年前的AI模式在所有这些任务上都会失败。
但对于我来说，Qwen结果和其他使用Python的实验并不属于直接读取重量的实验，一种既能实现“从零开始对小游戏的授权权重”又能实现“改进大模型在样本任务上的性能”的单一技术将意味着一种深度能力。 如果你测试一种单一的技巧，你用这种方法演示了一种“gestalt”的权重(它可以从零开始写出来)，但这里完全不是这样，因为它不是一个单一的技术。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Stage magicians often do the same trick multiple ways. One iteration uses a gimmick prop, the next one shows you the prop isn’t gimmick but that one palms the object, the third time was a forced choice etc. When you watch the performance you assume there’s only one technique, so each time they do it they seem to debunk your hypothesis about how it must be working.
This is pretty much how the experiments in the ‘Astra can read and write weights to improve Qwen’ post work. The post’s abstract says:
Just by looking at the raw weights, GPT-6 Astra can tell what a network was trained to do. And without any environment feedback, it has no trouble writing, from scratch, small game-playing networks that outplay networks trained with PPO for 100 million steps.
Likewise, asked to make Qwen3.5-0.8B better at math without gradients, Astra identifies an issue in its weights causing reasoning loops that Qwen’s own model card warns about, and fixes this by natively editing 0.016% of the weights. With the same sampling settings, this weight correction raises MATH-500 from 7.9% to 48.1% and generalizes out of the box to instruction following, raising IFEval by 3x.
If you read that, would you assume that the Qwen experiment worked ‘just by looking at the raw weights’? Well it doesn’t. In fact even the “small gameplaying network” experiment doesn’t work “just by looking at the raw weights”. There’s experiments that look at raw weights, and experiments that improve a game-playing network, and an experiment that improves Qwen at maths. But those latter two aren’t just looking at the raw weights.
Here’s a summary of what the post actually does and how the techniques work.
The “just by looking at the raw weights” setting (raw floats in context, no tools, no running the network) only appears in experiments 1, 2 and 4, on networks of 17K–172K weights. Every experiment involving a real language model uses Python, and the Qwen result, which is the headline one, also gets to run the model on up to 50 batches of practice problems.
What the Qwen result actually shows is that Astra can be prompted to devise and perform tasks that a human ML engineer could do. Human ML engineers write snippets of code that analyse or even modify network weights “directly” all the time. Figuring out how to do this productively is not nothing, and AI models of a year or two ago would have failed at all these tasks.
But to me the Qwen result and the other experiments that use Python don’t belong in the same post or paper as the experiments that read the weights directly. A single technique that could achieve both, “Author weights for a small game from scratch” and “Improve the performance of a large model on a sample task” would imply a deep capability. If you test a single technique that way you’ve demonstrated a sort of ‘gestalt’ of the weights (it can write them out from scratch) and also that it understands a large existing network. But that’s not at all true here, because it’s not a single technique.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.