DePT: Decomposed Prompt Tuning for Parameter-Efficient Fine-tuning

arXiv:2309.05173v5 Announce Type: replace-cross
Abstract: Prompt tuning (PT), where a small amount of trainable soft (continuous) prompt vectors is affixed to the input of language models (LM), has shown promising results across various tasks and models for parameter-efficient fine-tuning (PEFT). PT stands out from other PEFT approaches because it maintains competitive performance with fewer trainable parameters and does not drastically scale up its parameters as the model size expands. However, PT introduces additional soft prompt tokens, leading to longer input …

abstract arxiv continuous cs.ai cs.cl cs.cv cs.lg fine-tuning language language models peft performance prompt prompt tuning small tasks type vectors