TARDiS : Text Augmentation for Refining Diversity and Separability

TL;DR

TARDiS is a novel LLM-based text augmentation method that enhances diversity and class separability in few-shot text classification by introducing new generation and alignment techniques.

Contribution

It proposes a comprehensive two-stage augmentation framework with class-specific prompts and class adaptation, improving over existing LLM-based augmentation methods.

Findings

Outperforms state-of-the-art LLM-based TA methods in few-shot tasks.

Enhances diversity and separability of generated examples.

Provides detailed analysis of each augmentation stage.

Abstract

Text augmentation (TA) is a critical technique for text classification, especially in few-shot settings. This paper introduces a novel LLM-based TA method, TARDiS, to address challenges inherent in the generation and alignment stages of two-stage TA methods. For the generation stage, we propose two generation processes, SEG and CEG, incorporating multiple class-specific prompts to enhance diversity and separability. For the alignment stage, we introduce a class adaptation (CA) method to ensure that generated examples align with their target classes through verification and modification. Experimental results demonstrate TARDiS's effectiveness, outperforming state-of-the-art LLM-based TA methods in various few-shot text classification tasks. An in-depth analysis confirms the detailed behaviors at each stage.