今天，我们将一同探寻，AI如何通过理解“点的流动”来获得物理直觉，又为何会掉进人类语言的“隐喻陷阱”。我们还会深入AI的大脑，看看它的知识是如何生长又被遗忘的，并学习一种给AI做“大脑针灸”的调教术，治好它的“固执病”。最后，我们将揭秘一项最新研究，看看AI是如何被教会用“道德”这把尺子，去读懂网络上的“阴阳怪气”的。

00:00:34 机器人如何获得“物理直觉”？

00:05:05 你以为AI很理性？其实它活在比喻里

00:10:54 AI的大脑里，知识是怎么“长”出来又“丢”掉的？

00:16:36 AI调教术，给模型做一次“大脑针灸”

00:22:42 教AI读懂“阴阳怪气”，靠的是什么？

本期介绍的几篇论文：

[RO] PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation  

[Stanford University & NVIDIA]  

https://arxiv.org/abs/2601.03782 

---

[CL] Metaphors are a Source of Cross-Domain Misalignment of Large Reasoning Models  

[The University of New South Wales & CSIRO Data61]  

https://arxiv.org/abs/2601.03388 

---

[CL] How Do Large Language Models Learn Concepts During Continual Pre-Training?  

[UC Davis & Virginia Tech & UCLA]  

https://arxiv.org/abs/2601.03570 

---

[CL] ContextFocus: Activation Steering for Contextual Faithfulness in Large Language Models  

[Adobe Research, India]  

https://arxiv.org/abs/2601.04131 

---

[CL] Self-Explaining Hate Speech Detection with Moral Rationales  

[University of São Paulo & University of Southern California & Saarland University]  

https://arxiv.org/abs/2601.03481