Sign up to save your podcasts
Or
Share DoRA: Weight-Decomposed Low-Rank Adaptation
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
DoRA: Weight-Decomposed Low-Rank Adaptation
An upgrade over LoRA for parameter-efficient fine-tuning, enabling better performance in LLMs by decomposing weights into magnitude and direction components.
View all episodes
By Shaoqing TanAn upgrade over LoRA for parameter-efficient fine-tuning, enabling better performance in LLMs by decomposing weights into magnitude and direction components.