Can tiny adapters become personal AI assistants at scale?

Mind Lab, :, Song Cao, Vic Cao, Kaijie Chen, Bunny Fan, Hera Feng, Huan Feng, Arthur Fu, Jun Gao, Hongquan Gu, Aaron Guan, Mutian Hong, Hailee Hou, Peixuan Hua, Charles Huang, Miles Jiang, Nora Jiang, Yuyi Jiang, Autumn Jin, Fancy Kong, Kyrie Lei, Alexy Li, Dawn Li, Ray Li, Theo Li, Wenhao Li, Jiayi Lin, Domini Liu, Heshan Liu, Kairus Liu, Logan Liu, Maeve Luo, Runism Lv, Pony Ma, Verity Niu, Anson Qiu, Vincent Wang, Maxwell Yao, Regis Ye, Wenlin Ye, Yanying Ye, Josh Ying, Danney Zeng, Salmon Zhan, Anya Zhang, Ruijia Zhang, Shiyang Zhang, Sueky Zhang, Ya Zhang, Wei Zhao, Ada Zhou, Sizer Zhou, Xinyue Zhu, Murphy Zhuang

Instead of viewing parameter-efficient fine-tuning as merely a cheaper alternative to full training, researchers propose treating small adapters as persistent personal state—storing user preferences, skills, and memory on top of shared foundation models. They explore three scaling directions: stronger base models making small updates more useful, minimal adapter sizes before reliability breaks, and managing millions of concurrent personalized instances. MinT, their infrastructure system, handles adapter versioning, evaluation, and deployment. This reframes PEFT from cost-cutting hack to core substrate for scalable personal AI.