我们总说AI有知识,但你想过吗,AI的知识该如何称重、如何存储、又该如何溯源?更进一步,AI能否自我修炼、检查作业,它吃的“数据大餐”又藏着怎样的“秘密食谱”?今天,我们就从五篇最新的论文出发,一起探索AI知识世界的台前与幕后。
给AI模型称重,我们终于有了一杆新秤
AI的“记忆宫殿”是如何搭建的?
AI的自我修炼,如何从“检查作业”中获得智慧
AI世界的“亲子鉴定”技术
AI的“隐藏食谱”,为什么数据配比比数量更重要?
本期介绍的几篇论文:
[LG] Requential Coding: Pushing the Limits of Model Compression with Self-Generated Training Data
[New York University & CMU]
---
[LG] MLPs are Hebbians: Constructing Efficient Fact-Storing MLPs for Transformers
[Stanford University]
---
[AI] SVR-R1: Bootstrapping Multi-modal Reasoning with Self-verification in Reinforcement Learning
[University of Illinois Urbana-Champaign & Meta]
---
[LG] Reference-Based Distillation Detection in LLMs
[UC Berkeley]
---
[LG] Domain-Aware Scaling Laws Uncover Data Synergy
[MIT & Microsoft Research]
![[人人能懂AI前沿] 从检查作业、搭建记忆宫殿到寻找隐藏食谱](https://image.xyzcdn.net/FqWpK8fpivLboaqBbRHUe_BCOvxu.png@small)