或许，真正的成长，就始于你收回期待，敢于让某些人失望的那一刻。

00:00:38 AI进化论：从“打工人”到“CEO” 00:03:57 驯服AI的秘密：一句话，放在哪儿说效果大不同 00:07:58 你看到的，就是真相吗？——AI给我们的一个新警告 00:12:40 数据稀缺的时代，如何拼出一张完整的世界地图？ 00:17:23 AI正在偷偷学心理学，但好像学偏了 本期介绍的五篇论文： [LG] MetaAgent: Automatically Constructing Multi-Agent Systems Based on Finite State Machines [University of Wisconsin - Madison] https://arxiv.org/abs/2507.22606 --- [CL] Where to show Demos in Your Prompt: A Positional Bias of In-Context Learning [University of Maryland] https://arxiv.org/abs/2507.22887 --- [LG] Representation biases: will we achieve complete understanding by analyzing representations? [Google DeepMind] https://arxiv.org/abs/2507.22216 --- [LG] AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data [Google DeepMind] https://arxiv.org/abs/2507.22291 --- [LG] The Incomplete Bridge: How AI Research (Mis)Engages with Psychology [Johns Hopkins University & Rice University & Microsoft Research Asia] https://arxiv.org/abs/2507.22847

我们花了无数精力去设计能战胜人类的AI，却很少思考，如何为自己设计一个能战胜‘人性’的系统。

00:00:33 高手与普通人的差距，在于“记忆预算”的分配 00:04:16 AI当“牛顿”：我们如何找到万物生长的公式？ 00:08:26 让AI不止听话，更要会提问 00:12:09 AI 思考的艺术：如何做到又快又好？ 00:18:06 AI识人心：20盘棋，就“看穿”了你 本期介绍的五篇论文： [LG] Capacity-Constrained Continual Learning [Google DeepMind] https://arxiv.org/abs/2507.21479 --- [LG] EvoSLD: Automated Neural Scaling Law Discovery With Large Language Models [Peking University & Tsinghua University] https://arxiv.org/abs/2507.21184 --- [LG] Teaching Language Models To Gather Information Proactively [Microsoft] https://arxiv.org/abs/2507.21389 --- [LG] TriangleMix: A Lossless and Efficient Attention Pattern for Long Context Prefilling [Microsoft Research] https://arxiv.org/abs/2507.21526 --- [LG] Learning to Imitate with Less: Efficient Individual Behavior Modeling in Chess [University of Toronto] https://arxiv.org/abs/2507.21488

关于成功，总有一个变量，是连最强的AI都难以计算的，这个变量，就藏在我们今天的故事里。

00:00:34 让AI更“有谱儿”：不止一条路通罗马 00:05:21 如何让AI更聪明？一个“求稳”的智慧 00:09:14 造车新智慧：如何用“搬沙子”的办法，算出最省油的外形？ 00:12:31 AI也会“路径依赖”？一个简单动作，让它“老树发新芽” 00:17:01 AI炼丹术：我们如何“教会”机器遵守化学规则？ 00:21:50 AI进化论：从“死记硬背”到“自我成长” 本期介绍的几篇论文： [LG] Flow Matching Policy Gradients [UC Berkeley] https://arxiv.org/abs/2507.21053 --- [CL] Geometric-Mean Policy Optimization [Microsoft Research] https://arxiv.org/abs/2507.20673 --- [LG] Geometric Operator Learning with Optimal Transport [California Institute of Technology & Nvidia] https://arxiv.org/abs/2507.20065 --- [LG] What Can Grokking Teach Us About Learning Under Nonstationarity? [Google DeepMind] https://arxiv.org/abs/2507.20057 --- [LG] Enhancing Materials Discovery with Valence Constrained Design in Generative Modeling [MIT] https://arxiv.org/abs/2507.19799 --- [LG] A Survey of Self-Evolving Agents: On Path to Artificial Super Intelligence https://arxiv.org/abs/2507.21046

当你能清晰地分辨出什么是值得你全身心投入的‘信号’，并且有能力为它创造一个‘无菌舱’去深度运算时，你就不再是时间的囚徒，而是与时间共舞的伙伴。

00:00:29 AI 进阶之路：不造轮子，而是给高手装上预知未来的眼睛 00:04:38 AI进化新思路：不说人话，怎么教得会？ 00:09:24 你的 App 突然崩了？别急，AI 程序员正在赶来修复的路上 00:15:21 AI 也会看走眼？我们如何教它练就一双“火眼金睛” 00:19:21 如何教AI学会“举一反三”？ 本期介绍的五篇论文： [CV] Back to the Features: DINO as a Foundation for Video World Models [Meta FAIR] https://arxiv.org/abs/2507.19468 --- [CL] GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning [UC Berkeley & Stanford University] https://arxiv.org/abs/2507.19457 --- [LG] Agentic Program Repair from Test Failures at Scale: A Neuro-symbolic approach with static analysis and test execution feedback [Meta] https://arxiv.org/abs/2507.18755 --- [CL] PrismRAG: Boosting RAG Factuality with Distractor Resilience and Strategized Reasoning [Meta Reality Labs] https://arxiv.org/abs/2507.18857 --- [LG] Scale-Consistent Learning for Partial Differential Equations [Caltech & Nvidia] https://arxiv.org/abs/2507.18813

我们一直在做加法，却忘了成功的另一半，是“减法”。

00:00:32 你的夸奖，正在“毒害”AI 00:05:22 数据大扫除：不止是扔垃圾，更是换风格 00:10:55 AI的“世界观”：它如何从零开始看懂现实？ 00:15:46 AI的“省钱攻略”：如何花小钱办大事？ 00:20:27 喂养AI的新艺术：从“吃什么”到“怎么吃” 本期介绍的无篇文章： [LG] Off-Policy Corrected Reward Modeling for Reinforcement Learning from Human Feedback [The University of Tokyo and RIKEN AIP] https://arxiv.org/abs/2507.15507 --- [LG] Distributional Unlearning: Forgetting Distributions, Not Just Samples [EPFL & Stanford University] https://arxiv.org/abs/2507.15112 --- [LG] Skill Learning via Policy Diversity Yields Identifiable Representations for Reinforcement Learning [Max Planck Institute for Intelligent Systems & University of Tübingen] https://arxiv.org/abs/2507.14748 --- [CL] Towards Compute-Optimal Many-Shot In-Context Learning [Google Cloud AI Research] https://arxiv.org/abs/2507.16217 --- [LG] LLM Data Selection and Utilization via Dynamic Bi-level Optimization [University of Chinese Academy of Sciences & Huawei Noah’s Ark Lab] https://arxiv.org/abs/2507.16178

“心有所信，方能行远。” 真正的“远”，不是抵达预设的终点，而是当你的信念被全世界质疑时，你依然敢于向你的造物，问出一个天真的问题，并最终，从它那里，听到了一声来自深渊的回响。

00:00:35 AI进化论：当机器开始自己设计自己 00:04:59 AI训练场上的新规则：别纠结错别字，看的是整篇文章 00:09:53 AI当医生？先等等，咱们换个活法儿 00:15:29 给AI做体检：我们能不花钱就看出模型好坏吗？ 00:20:35 AI没书读了怎么办？一个“笨”方法里的新智慧 本期节目介绍的五篇论文： [LG] AlphaGo Moment for Model Architecture Discovery [Shanghai Jiao Tong University & SII] https://arxiv.org/abs/2507.18074 --- [LG] Group Sequence Policy Optimization [Qwen Team] https://arxiv.org/abs/2507.18071 --- [LG] Towards physician-centered oversight of conversational diagnostic AI [Google DeepMind & Google Research] https://arxiv.org/abs/2507.15743 --- [LG] SETOL: A Semi-Empirical Theory of (Deep) Learning [Calculation Consulting & Onyx Point Systems] https://arxiv.org/abs/2507.179 --- [LG] Diffusion Beats Autoregressive in Data-Constrained Settings [CMU & Lambda] https://arxiv.org/abs/2507.15857