[人人能懂] 从“数据微整形”到“量体裁衣”

00:00:37 给数据做“微整形”：如何把猫的“样子”变成狗？00:05:01 从一堆曲线里，看出“命”来00:09:53 怎么给AI模型打一针“思想疫苗”？00:13:53 AI的“少年时代”：为什么有些超能力，学会了又忘了？00:18:30 AI界的“量体裁衣”：如何给数据穿上最合身的“衣服”？本期介绍的几篇论文：[LG] Flowing Datasets with Wasserstein over Wasserstein Gradient Flows  [ENSAE & Universite Paris-Saclay]  https://openreview.net/forum?id=I1OHPb4zWo  ---[LG] Learning with Expected Signatures: Theory and Applications  [Imperial College London]  https://openreview.net/forum?id=yDTwamN4LQ  ---[LG] Model Immunization from a Condition Number Perspective  [Purdue University]  https://openreview.net/forum?id=uitj69FqD5  ---[LG] Strategy Coopetition Explains the Emergence and Transience of In-Context Learning  [University College London & Anthropic AI & University of Oxford]  https://openreview.net/forum?id=esBoQFmD7v  ---[LG] AutoGFM: Automated Graph Foundation Model with Adaptive Architecture Customization  [Tsinghua University]  https://openreview.net/forum?id=fCPB0qRJT2  在小宇宙查看该单集文稿