让AI直接改模型权重,先得把模型练「好改」
大模型能直接改另一个神经网络的权重,这件事已经有人做到了;但没人问过:被改的那个网络,能不能事先练得「好改」。这篇给了第一份系统评测:让 5 个大模型去改卷积网络和视觉 Transformer,在四个数据集上做重标注和纠错,一共跑了 2600 多次编辑,每次都在没见过的图上打分,看改没改成、有没有副作用、原精度掉没掉。结论是:标准训练出来的网络很难改,而他们提出的 Edit-Ready Training 先让网络「排练」一遍大模型会做的那种闭式权重修改,改起来成功率明显更高,GPT-6 Astra 和 Claude Opus 5.5 在它上面的平均分也最高。它不是你明天能用上的东西。
📄 原文摘要(英文)
It has been demonstrated that, in certain settings, large language models can perform weight-space programming to directly modify the weights of neural networks. However, it is unknown how dedicated training of these networks can contribute to the successes of weight-space programming. We present a benchmark of LLM-driven weight-space programming on vision models, in which five LLMs edit convolutional networks and vision transformers on relabeling and error-correction tasks on Waterbirds, CIFAR-100, PACS and Entity-30. We report more than 2,600 such edit processes, whereby each edit is scored on held-out images for success, side effects and accuracy loss. We compare standard training with label smoothing, gradient-based editable training, and our Edit-Ready Training (ERT), a novel method that rehearses the direct, closed-form weight edits LLMs perform. We show that preparing networks for editing raises the success of LLM edits substantially where standard networks are hard to edit. ERT further achieves the highest mean score for GPT-6 Astra and Claude Opus 5.5 across all evaluated baselines. We will release our LLM testing framework alongside all tasks, editing trajectories and scores. Project page: https://viscom.uni-ulm.de/wsp