Guiding Generative Language Models for Data Augmentation in Few-Shot Text Classification
Guiding Generative Language Models for Data Augmentation in Few-Shot Text Classification
复制标题
DOI：
--
复制
发表时间：
2021-11
期刊：
影响因子：
--
通讯作者：
A. Edwards;Asahi Ushio;José Camacho-Collados;Hélène de Ribaupierre;A. Preece
中科院分区：
文献类型：
--
作者：
A. Edwards;Asahi Ushio;José Camacho-Collados;Hélène de Ribaupierre;A. Preece
数据增强技术被广泛用于通过解决类不平衡问题和数据稀疏性来提高机器学习模型的性能。最先进的生成语言模型已被证明在不同的NLP任务中提供显著的收益。然而，它们在文本分类任务的数据增强方面的适用性还没有得到充分的探索，特别是在专门的领域。在本文中，我们利用GPT-2 （Radford等人，2019）来生成人工训练实例，以提高分类性能。我们的目的是分析种子训练样本的选择过程对gpt生成样本的质量以及分类器性能的影响。我们提出了一种人在环的方法来选择种子样本。此外，我们将该方法与其他利用特定领域（如人类创造的类层次结构和名词短语的存在）特征的种子选择策略进行了比较。我们的结果表明，在少数标签实例中微调GPT-2会导致一致的分类改进，并优于竞争基准。在这项工作中开发的种子选择策略导致显著改进随机种子选择的专门领域。我们表明，通过领域专家选择来指导文本生成可以导致进一步的改进，这为结合生成模型和主动学习开辟了有趣的研究途径。
Data augmentation techniques are widely used for enhancing the performance of machine learning models by tackling class imbalance issues and data sparsity. State-of-the-art generative language models have been shown to provide significant gains across different NLP tasks. However, their applicability to data augmentation for text classification tasks in few-shot settings have not been fully explored, especially for specialised domains. In this paper, we leverage GPT-2 (Radford et al, 2019) for generating artificial training instances in order to improve classification performance. Our aim is to analyse the impact the selection process of seed training examples has over the quality of GPT-generated samples and consequently the classifier performance. We propose a human-in-the-loop approach for selecting seed samples. Further, we compare the approach to other seed selection strategies that exploit the characteristics of specialised domains such as human-created class hierarchical structure and the presence of noun phrases. Our results show that fine-tuning GPT-2 in a handful of label instances leads to consistent classification improvements and outperform competitive baselines. The seed selection strategies developed in this work lead to significant improvements over random seed selection for specialised domains. We show that guiding text generation through domain expert selection can lead to further improvements, which opens up interesting research avenues for combining generative models and active learning.