LoRA

mainstreamstable

A parameter-efficient adaptation method that trains low-rank updates while keeping base model weights frozen.

雷达信号

定义

Low-Rank Adaptation (LoRA) adapts a pretrained model by learning small low-rank matrices instead of updating all original weights. The method reduces the trainable parameter count and associated training memory. A LoRA adapter represents a task-specific update; it is not necessarily a standalone model.

来源

Research paper first submitted to arXiv on 2021-06-17.

语境

Useful for comparing fine-tuning techniques and downloadable adapters. Compatibility depends on the base architecture and the weights the adapter was trained against. Catalog note: this is a source-verified reference entry, not a measured current-trend signal. No search-volume or growth claim is made.

来源

LoRA: Low-Rank Adaptation of Large Language Models

Trend score: 0
First seen: 2021-06-17
Last seen: 2026-10-24

看今天的表

专业版打开当天的 prioritize / watch / skip 表，包括这里还不能公开的新名字。