P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks

TL;DR

P-Tuning v2 demonstrates that prompt tuning can match fine-tuning performance across various model sizes and NLU tasks, offering a more parameter-efficient alternative with broad applicability.

Contribution

The paper introduces P-Tuning v2, a universal prompt tuning method optimized for NLU that achieves comparable results to fine-tuning across diverse models and tasks.

Findings

Prompt tuning can be universally effective across model scales and tasks.

P-Tuning v2 matches fine-tuning performance with only 0.1%-3% of parameters tuned.

The method serves as a strong, simple baseline for future research.

Abstract

Prompt tuning, which only tunes continuous prompts with a frozen language model, substantially reduces per-task storage and memory usage at training. However, in the context of NLU, prior work reveals that prompt tuning does not perform well for normal-sized pretrained models. We also find that existing methods of prompt tuning cannot handle hard sequence labeling tasks, indicating a lack of universality. We present a novel empirical finding that properly optimized prompt tuning can be universally effective across a wide range of model scales and NLU tasks. It matches the performance of finetuning while having only 0.1%-3% tuned parameters. Our method P-Tuning v2 is an implementation of Deep Prompt Tuning \cite{li2021prefix,qin2021learning} optimized and adapted for NLU. Given the universality and simplicity of P-Tuning v2, we believe it can serve as an alternative to finetuning and a strong baseline for future research.Our code and data are released at https://github.com/THUDM/P-tuning-v2.