arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.01431cs.CL
收藏

STAD:低资源关系抽取的模糊数据自训练

STAD: Self-Training with Ambiguous Data for Low-Resource Relation Extraction

Soochow University(苏州大学);Tencent AI Lab(腾讯人工智能实验室);China Mobile Online Marketing and Services Center(中国移动在线营销服务中心)

Junjie Yu, Xing Wang, Jiangjiang Zhao, Chunjie Yang, Wenliang Chen

摘要:

我们提出了一个简单而有效的自我训练方法,命名为STAD,低资源的关系抽取。该方法首先根据教师模型预测的概率将自动标注的实例分为两组:置信实例和不确定实例。与以往大多数研究中只使用有把握的样本进行自我训练不同,本文利用了不确定的样本。为此,我们提出了一种从不确定实例中识别歧义但有用的实例的方法,然后针对每个歧义实例将关系分为候选标签集和负标签集。接下来,我们提出了一个集负训练方法的负标签集的歧义实例和一个积极的训练方法的置信实例。最后,提出了一种联合训练的方法来建立最终的关系抽取系统的所有数据。在两个广泛使用的数据集SemEval 2010 Task-8和Re-TACRED上进行的低资源设置的实验结果表明,与几个有竞争力的自训练系统相比,这种新的自训练方法确实实现了显着和一致的改进。代码可在https://github.com/jjyunlp/STAD上公开获取

英文摘要:

We present a simple yet effective self-training approach, named as STAD, for low-resource relation extraction. The approach first classifies the auto-annotated instances into two groups: confident instances and uncertain instances, according to the probabilities predicted by a teacher model. In contrast to most previous studies, which mainly only use the confident instances for self-training, we make use of the uncertain instances. To this end, we propose a method to identify ambiguous but useful instances from the uncertain instances and then divide the relations into candidate-label set and negative-label set for each ambiguous instance. Next, we propose a set-negative training method on the negative-label sets for the ambiguous instances and a positive training method for the confident instances. Finally, a joint-training method is proposed to build the final relation extraction system on all data. Experimental results on two widely used datasets SemEval2010 Task-8 and Re-TACRED with low-resource settings demonstrate that this new self-training approach indeed achieves significant and consistent improvements when comparing to several competitive self-training systems. Code is publicly available at https://github.com/jjyunlp/STAD

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑