A Lightweight Multi Aspect Controlled Text Generation Solution For Large Language Models

TL;DR

This paper introduces a lightweight data augmentation approach to enhance large language models' ability to perform multi-aspect controllable text generation, addressing dataset biases and correlations for improved task performance.

Contribution

It proposes a novel data augmentation pipeline that improves LLMs' MCTG capabilities without complex model modifications, focusing on bias reduction and aspect control.

Findings

20% increase in accuracy for MCTG tasks

Reduced aspect correlations in augmented datasets

Enhanced adaptability of LLMs to controllable generation

Abstract

Large language models (LLMs) show remarkable abilities with instruction tuning. However, they fail to achieve ideal tasks when lacking high-quality instruction tuning data on target tasks. Multi-Aspect Controllable Text Generation (MCTG) is a representative task for this dilemma, where aspect datasets are usually biased and correlated. Existing work exploits additional model structures and strategies for solutions, limiting adaptability to LLMs. To activate MCTG ability of LLMs, we propose a lightweight MCTG pipeline based on data augmentation. We analyze bias and correlations in traditional datasets, and address these concerns with augmented control attributes and sentences. Augmented datasets are feasible for instruction tuning. In our experiments, LLMs perform better in MCTG after data augmentation, with a 20% accuracy rise and less aspect correlations.