[人人能懂] AI心智架构：从记忆、偏见到思考方式的全面升级

00:01:37 如何让你的机器人管家，不再健忘？00:05:43 你家的AI，可能有个“邪恶双胞胎”00:09:30 AI：一面照见我们自己的镜子00:13:20 AI训练场上的“复古”新潮流00:18:14 AI的“复盘”神功：如何想得更深？00:21:26 驯养AI：我们是如何让机器“猜”出你心思的？本期介绍的几篇论文：[RO] Enter the Mind Palace: Reasoning and Planning for Long-term Active Embodied Question Answering  [Field AI]  https://arxiv.org/abs/2507.12846  ---[CL] Jailbreak-Tuning: Models Efficiently Learn Jailbreak Susceptibility  [FAR.AI & Mila]  https://arxiv.org/abs/2507.11630  ---[CL] Emergence of Hierarchical Emotion Organization in Large Language Models  [Harvard University]  https://arxiv.org/abs/2507.10599  ---[LG] ShiQ: Bringing back Bellman to LLMs  [Cohere]  https://arxiv.org/abs/2505.11081  ---[LG] AbbIE: Autoregressive Block-Based Iterative Encoder for Efficient Sequence Modeling  [University of Cambridge]  https://arxiv.org/abs/2507.08567  ---[LG] Inverse Reinforcement Learning Meets Large Language Model Post-Training: Basics, Advances, and Opportunities  [Not explicitly stated, likely academic tutorial paper]  https://arxiv.org/abs/2507.13158  在小宇宙查看该单集文稿