按难易程度分批训练，提升中文语法纠错模型效果
Loss-Aware Curriculum Learning for Chinese Grammatical Error Correction
- 根据纠错难度分级，从易到难逐步训练样本
- 在多个数据集上显著提升纠错准确率
- 适合需要高效训练语法纠错模型的研究者
中文语法错误纠正（CGEC）旨在检测并修正输入中文句子中的错误。近年来，预训练语言模型（PLMs）被用于提升性能。然而，现有方法忽略了不同样本间纠错难度的差异，对所有样本一视同仁，增加了模型学习的难度。为此，我们提出一种多粒度课程学习（CL）框架。首先计算各样本的纠错难度，并按从易到难的顺序逐批输入模型；随后采用实例级课程学习，通过调节损失函数，引导模型自动朝合适方向优化。大量实验及多数据集分析验证了该方法的有效性。
原文摘要 · Abstract (English)
Chinese grammatical error correction (CGEC) aims to detect and correct errors in the input Chinese sentences. Recently, Pre-trained Language Models (PLMS) have been employed to improve the performance. However, current approaches ignore that correction difficulty varies across different instances and treat these samples equally, enhancing the challenge of model learning. To address this problem, we propose a multi-granularity Curriculum Learning (CL) framework. Specifically, we first calculate the correction difficulty of these samples and feed them into the model from easy to hard batch by batch. Then Instance-Level CL is employed to help the model optimize in the appropriate direction automatically by regulating the loss function. Extensive experimental results and comprehensive analyses of various datasets prove the effectiveness of our method.
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。