00:00:30 给AI请个“速记员”，效率提升30倍

00:04:57 AI为啥爱“猜答案”？这事儿得从考试说起

00:10:14 高手过招：为什么说“三次”探索胜过“两次”？

00:15:23 AI训练场上的老司机，如何做到油门收放自如？

00:19:24 不懂代码？没关系，AI帮你组建一个技术团队

本期介绍的几篇论文：

[CL] REFRAG: Rethinking RAG based Decoding  

[Meta Superintelligence Labs & National University of Singapore]  

https://arxiv.org/abs/2509.01092 

---

[LG] Why Language Models Hallucinate  

[OpenAI]  

https://cdn.openai.com/pdf/d04913be-3f6f-4d2b-b283-ff432ef4aaa5/why-language-models-hallucinate.pdf 

---

[LG] When three experiments are better than two: Avoiding intractable correlated aleatoric uncertainty by leveraging a novel bias--variance tradeoff  

[Relation]  

https://arxiv.org/abs/2509.04363 

---

[LG] AdaGrad Meets Muon: Adaptive Stepsizes for Orthogonal Updates  

[University of California, Los Angeles]  

https://arxiv.org/abs/2509.02981 

---

[LG] AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML  

[DeepAuto.ai]  

https://arxiv.org/abs/2410.02958