Primary sourcearXiv cs.LG
View sourceEvent ·Multi-source
WeightLoop框架让GPT-6 Astra用实验反馈直接改写模型权重
arXiv 论文介绍 WeightLoop 框架,属于权重空间编程(WSP)与递归自我改进(RSI)的结合。GPT-6 Astra 读取任务规范和模型实现后,生成一段有边界的可复用参数编辑程序。新测量数据到来时,程序找出预测偏差,用观测结果验证候选改动,再写回原模型。编辑器固定不变、预测器持续演化,更新过程不需要额外 LLM 调用或梯度重训练。离线回放中,定向编辑在 8 个端点预算设置中的 7 个提升了平均排序效用,对比随机编辑则是 8 个全部提升,并在 3 个序列任务上提高了平均最终发现数。
Current read
让模型自己看着实验数据改自己的权重,不用重训练。用 GPT-6 Astra 做的,回放里 8 组测试赢了 7 组,思路挺新。
11 sources21° AI heatLast updated 2026/11/5 02:06:51
Evidence
11 sourcesRelated source 1Simon Willison’s Weblog
View sourceRelated source 2李继刚
View sourceRelated source 3歸藏(guizang.ai)
View sourceRelated source 4IT之家
View sourceRelated source 5DAIR.AI
View sourceRelated source 6Latent.Space
View sourceRelated source 7Bindu Reddy
View sourceRelated source 8rohanpaul_ai
View sourceRelated source 9The Rundown AI
View sourceRelated source 10Decoder
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.