论文标题
NERDA-CON：扩展用于持续学习的NER模型 - 整合不同的任务并更新分配变化
NERDA-Con: Extending NER models for Continual Learning -- Integrating Distinct Tasks and Updating Distribution Shifts
论文作者
论文摘要
随着生物医学信息提取管道和社交媒体分析等领域的应用增加，名为实体识别（NER）已成为知识提取的必不可少的工具。但是，随着语言结构和词汇的逐渐变化，Ners遭受了分配的转变困扰，因此如果不重新训练，它们都会变得多余或不那么有利可图。基于大型语言模型（LLM）从头开始重新训练NER，从而对新获得的数据造成了经济缺点。相反，仅使用新获得的数据进行重新训练将导致灾难性忘记先前获得的知识。因此，我们提出了NERDA-CON，这是一种通过将弹性重量巩固（EWC）概念纳入NER微调NERDA管道中，用于训练LLM碱基的管道。正如我们认为我们的工作具有含义是在持续学习和NER的管道中使用的，我们开放式代码，并在https://github.com/supritivivivijay/nerda-con/nerda-con/nerda-con and https:/pypy.org/pypi.org/prapter/nerdaon/nerdaondandandandaon/nerdandaon/https:/supritivivijay/nerda-con/nerda-con/nerda-con/nerda-con/nerda-con/nerda-con-com.com/supritivivijay/nerda-com.com/supritivivijay/supritivivijay/nerda一下。
With increasing applications in areas such as biomedical information extraction pipelines and social media analytics, Named Entity Recognition (NER) has become an indispensable tool for knowledge extraction. However, with the gradual shift in language structure and vocabulary, NERs are plagued with distribution shifts, making them redundant or not as profitable without re-training. Re-training NERs based on Large Language Models (LLMs) from scratch over newly acquired data poses economic disadvantages. In contrast, re-training only with newly acquired data will result in Catastrophic Forgetting of previously acquired knowledge. Therefore, we propose NERDA-Con, a pipeline for training NERs with LLM bases by incorporating the concept of Elastic Weight Consolidation (EWC) into the NER fine-tuning NERDA pipeline. As we believe our work has implications to be utilized in the pipeline of continual learning and NER, we open-source our code as well as provide the fine-tuning library of the same name NERDA-Con at https://github.com/SupritiVijay/NERDA-Con and https://pypi.org/project/NERDA-Con/.
