PerPLM: Personalized Fine-tuning of Pretrained Language Models via Writer-specific Intermediate Learning and Prompts

TL;DR

This paper introduces PerPLM, a method for personalizing pretrained language models using writer-specific prompts and intermediate learning, improving text understanding by capturing individual writing traits without multiple model fine-tunings.

Contribution

It proposes a novel prompt-based personalization framework with intermediate learning to adapt PLMs to individual writers using only their plain text.

Findings

Prompt design significantly affects personalization effectiveness.

Intermediate learning enhances writer trait extraction.

Personalized prompts improve task accuracy across datasets.

Abstract

The meanings of words and phrases depend not only on where they are used (contexts) but also on who use them (writers). Pretrained language models (PLMs) are powerful tools for capturing context, but they are typically pretrained and fine-tuned for universal use across different writers. This study aims to improve the accuracy of text understanding tasks by personalizing the fine-tuning of PLMs for specific writers. We focus on a general setting where only the plain text from target writers are available for personalization. To avoid the cost of fine-tuning and storing multiple copies of PLMs for different users, we exhaustively explore using writer-specific prompts to personalize a unified PLM. Since the design and evaluation of these prompts is an underdeveloped area, we introduce and compare different types of prompts that are possible in our setting. To maximize the potential of prompt-based personalized fine-tuning, we propose a personalized intermediate learning based on masked language modeling to extract task-independent traits of writers' text. Our experiments, using multiple tasks, datasets, and PLMs, reveal the nature of different prompts and the effectiveness of our intermediate learning approach.