用五个微调模型在六个数据集上测试集成效果，发现集成能提升文本分类性能。
Ensembling Finetuned Language Models for Text Classification
- 基于五个微调模型的预测结果构建元数据集，评估不同集成策略
- 集成可显著提升微调模型在文本分类任务中的性能表现
- 为后续研究提供可复现的基准，适合关注模型集成的实践者
微调是将预训练模型适配到特定任务的常用方法，文本分类是其中重要应用之一。尽管神经网络集成常用于提升性能并提供可靠的不确定性估计，但针对文本分类任务中预训练模型的集成仍缺乏系统研究。本文构建了一个包含五个大型微调模型在六个数据集上的预测结果的元数据集，并报告了多种集成策略的实验结果。研究揭示了集成如何有效提升微调文本分类器的性能，为未来在该任务中推广集成方法提供了依据。
原文摘要 · Abstract (English)
Finetuning is a common practice widespread across different communities to adapt pretrained models to particular tasks. Text classification is one of these tasks for which many pretrained models are available. On the other hand, ensembles of neural networks are typically used to boost performance and provide reliable uncertainty estimates. However, ensembling pretrained models for text classification is not a well-studied avenue. In this paper, we present a metadataset with predictions from five large finetuned models on six datasets, and report results of different ensembling strategies from these predictions. Our results shed light on how ensembling can improve the performance of finetuned text classifiers and incentivize future adoption of ensembles in such tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。