#lora

• 2 Jun 2026 · Paper summary

  Paper summary: On the Scaling of PEFT: Towards Million Personal Models of Trillion Parameters

  This position-style study argues that small adapters on top of shared foundation models can serve as persistent personal state (preferences, skills, tool habits, memory-like updates), organised along Scale Up, Scale Down and Scale Out axes.

• 14 May 2026 · Paper summary

  Paper summary: MinT: Managed Infrastructure for Training and Serving Millions of LLMs

  MinT is infrastructure for training and serving many LoRA policies on a few shared base-model deployments, moving only adapters through rollout, update, evaluation and serving, scaled past 1T-parameter bases and toward million-adapter catalogues.

• 25 Feb 2026 · Study

  LLMs · Fine-tuning with LoRA

  LoRA freezes the pretrained weights and learns a low-rank update BA; a LoRA layer written from scratch, then PEFT applied to GPT-2 to count how few parameters are trained.