[人人能懂] 解锁AI新维度：思考、行动、创造、自知

00:00:30 AI界的“学霸”是怎么炼成的？00:04:45 你的下一个AI，为什么必须是个“行动派”？00:08:35 AI当上帝：从零开始创造一门语言00:13:09 如何给AI大模型“摸骨”，看透它的知识边界？本期介绍的四篇论文：[CL] UR²: Unify RAG and Reasoning through Reinforcement Learning  [Tsinghua University & Hebei University of Economics and Business]  https://arxiv.org/abs/2508.06165 ---[CL] GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models  [Zhipu AI & Tsinghua University]  https://arxiv.org/abs/2508.06471 ---[CL] ConlangCrafter: Constructing Languages with a Multi-Hop LLM Pipeline  [Tel Aviv University & UC Berkeley]  https://arxiv.org/abs/2508.06094 ---[CL] Efficient Knowledge Probing of Large Language Models by Adapting Pre-trained Embeddings  [Georgia Institute of Technology & MIT]  https://arxiv.org/abs/2508.06030 在小宇宙查看该单集文稿