Task-Adaptive Tokenization: Enhancing Long-Form Text Generation Efficacy in Mental Health and Beyond

TL;DR

This paper introduces task-adaptive tokenization, which customizes token segmentation for specific tasks, improving long-form text generation in mental health applications by reducing token usage and enhancing performance.

Contribution

The paper presents a novel task-adaptive tokenization method that samples variable segmentations and integrates task-specific tokens into pre-trained models, tailored for mental health text generation.

Findings

Significant performance improvements in psychological question-answering tasks.

Up to 60% reduction in token usage while maintaining quality.

Promising initial results with large language models.

Abstract

We propose task-adaptive tokenization as a way to adapt the generation pipeline to the specifics of a downstream task and enhance long-form generation in mental health. Inspired by insights from cognitive science, our task-adaptive tokenizer samples variable segmentations from multiple outcomes, with sampling probabilities optimized based on task-specific data. We introduce a strategy for building a specialized vocabulary and introduce a vocabulary merging protocol that allows for the integration of task-specific tokens into the pre-trained model's tokenization step. Through extensive experiments on psychological question-answering tasks in both Chinese and English, we find that our task-adaptive tokenization approach brings a significant improvement in generation performance while using up to 60% fewer tokens. Preliminary experiments point to promising results when using our tokenization approach with very large language models.