Primary source李继刚
View sourceEvent ·Multi-source
用 GPT-6 Astra 直接改 Qwen3.5-0.8B 权重，数学能力提升 6 倍
Rikiya Takehi 团队让 GPT-6 Astra 直接编辑 Qwen3.5-0.8B 的权重来提升数学能力。整个过程只改动了 0.016% 的权重，不使用梯度、不做训练，修复了 0.8B 模型的推理死循环。改动后数学能力最高提升 6 倍，指令遵循能力提升 3 倍。团队将这种方法称为 Weight-Space Programming（权重空间编程），认为前沿 LLM 已能原生读写神经网络权重。
Current read
有人让 GPT-6 Astra 不训练直接改 Qwen3.5-0.8B 的权重，数学提升 6 倍，思路很新鲜，帖子不长值得看看。
11 sources23° AI heatLast updated 2026/11/3 02:19:10
Evidence
11 sourcesRelated source 1orange.ai
View sourceRelated source 2Thomas Wolf
View sourceRelated source 3The Rundown AI
View sourceRelated source 4elvis
View sourceRelated source 5Jerry Liu
View sourceRelated source 6shao__meng
View sourceRelated source 7歸藏(guizang.ai)
View sourceRelated source 8IT之家
View sourceRelated source 9kimmonismus
View sourceRelated source 10rohanpaul_ai
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.