Chubby♨️· @kimmonismus · X·· 3 小时前AI 评分66
AI 导读
MIT 研究人员称，GPT-6 Astra 通过直接编辑 Qwen3.5-0.8B 仅 0.016% 的权重，把该模型的数学分数从 7.9% 提升到 48.1%，全程不依赖梯度训练。引用内容将这一方法称为 Weight-Space Programming，称其修复了 0.8B 模型的推理循环，数学能力最高提升 6 倍、指令遵循提升 3 倍。作者认为，这指向 AI 系统能够直接诊断并改进其他模型，是迈向自我改进 AI 的一步。
正文
Absolutely incredible: MIT researchers say GPT-6 Astra improved a small Qwen model’s math score from 7.9% to 48.1% by directly editing just 0.016% of its weights, without gradient-based training.
The result points toward AI systems that can diagnose and improve other models directly, a potential step toward self-improving AI.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: https://rikiyat.github.io/blog/weight-space-programming/在 X 查看被引用的帖子
来源：Chubby♨️ · x.com