<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[m0_53371353的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/m0_53371353</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; m0_53371353]]></copyright><item><title><![CDATA[损失test1]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/132866366</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/132866366</guid><author>m0_53371353</author><pubDate>Sun, 07 Apr 2024 18:24:52 +0800</pubDate><description><![CDATA[最近，随着预训练的基于transformer的语言模型的巨大成功和BERT一样，它们被广泛用于生成句子表示。一种直接的方法是利用[CLS]嵌入(Devlin等人，2019)或在类bert预训练语言模型的最后几层应用均值池(Reimers和Gurevych, 2019)。然而，未经进一步微调的预训练语言模型的句子嵌入不能很好地捕获句子的语义，有时甚至不如GloVe 等非上下文化嵌入。为了对句子之间的正负成对关系进行建模，我们首先需要生成句子表征，并将它们分组为正负对。然后将这些对输入到训练目标中进行优化。]]></description><category></category></item><item><title><![CDATA[代码相关python]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/130755323</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/130755323</guid><author>m0_53371353</author><pubDate>Tue, 11 Jul 2023 22:51:51 +0800</pubDate><description><![CDATA[python代码]]></description><category></category></item><item><title><![CDATA[bert相关]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/130713726</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/130713726</guid><author>m0_53371353</author><pubDate>Tue, 16 May 2023 21:56:51 +0800</pubDate><description><![CDATA[1、自注意力机制












]]></description><category></category></item><item><title><![CDATA[每天读一篇文献]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/130564544</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/130564544</guid><author>m0_53371353</author><pubDate>Mon, 08 May 2023 22:21:07 +0800</pubDate><description><![CDATA[本文提出了一种基于知识增强的预训练模型作为语义表示层，在特征提取层采用和改进的融合模型。实验表明，该模型在。在多标签分类任务中的数据不平衡问题中，利用实现模型更快收敛的目标。1、针对实体关系抽取流水线化的问题，采用基于序列标注的联合方法进行实体关系抽取。2、采用知识增强的ERNIE预训练模型进行文本语义表示。联合学习方法。]]></description><category></category></item><item><title><![CDATA[杂七杂八的知识]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/130565285</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/130565285</guid><author>m0_53371353</author><pubDate>Mon, 08 May 2023 19:07:12 +0800</pubDate><description><![CDATA[杂七杂八]]></description><category></category></item><item><title><![CDATA[每天读一篇文献15--CBLUE]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/130150602</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/130150602</guid><author>m0_53371353</author><pubDate>Fri, 14 Apr 2023 16:07:08 +0800</pubDate><description><![CDATA[论文学习]]></description><category></category></item><item><title><![CDATA[每天读一篇文献14--GraphRel: Modeling Text as Relational Graphs for Joint Entity andRelation Extraction]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/130123220</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/130123220</guid><author>m0_53371353</author><pubDate>Thu, 13 Apr 2023 10:22:42 +0800</pubDate><description><![CDATA[论文阅读]]></description><category></category></item><item><title><![CDATA[每天读一篇文献13--DGCL: Distance-wise and Graph Contrastive Learning for medicationrecommendation]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129750291</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129750291</guid><author>m0_53371353</author><pubDate>Fri, 24 Mar 2023 14:54:26 +0800</pubDate><description><![CDATA[DGCL: Distance-wise and Graph Contrastive Learning for medication
recommendation]]></description><category></category></item><item><title><![CDATA[每天读一篇文献12--A deep learning method to detect prescription and use disorder from electronic health]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129741452</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129741452</guid><author>m0_53371353</author><pubDate>Thu, 23 Mar 2023 23:31:57 +0800</pubDate><description><![CDATA[本研究的目的是开发一种结合深度学习的方法，利用结构化和非结构化EHR数据预测某类药物处方和use disorder。]]></description><category></category></item><item><title><![CDATA[每天读一篇文献11--Information Extraction from the Text Data onTraditional Chinese Medicine]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129720752</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129720752</guid><author>m0_53371353</author><pubDate>Wed, 22 Mar 2023 22:17:34 +0800</pubDate><description><![CDATA[Information Extraction from the Text Data onTraditional Chinese Medicine: A Review on Tasks, Challenges, and Methods from 2010 to 2021]]></description><category></category></item><item><title><![CDATA[每天读一篇文献10--Rediscovery of CNN’s Versatility for Text-based Encoding of Raw Electronic Health Records]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129657557</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129657557</guid><author>m0_53371353</author><pubDate>Tue, 21 Mar 2023 21:07:09 +0800</pubDate><description><![CDATA[Rediscovery of CNN’s Versatility for Text-based Encoding of Raw Electronic Health Records重新发现CNN对原始电子健康记录文本编码的多功能性]]></description><category></category></item><item><title><![CDATA[每天读一篇文献9--UniHPF : Universal Healthcare Predictive Framework with Zero Domain Knowledge]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129671217</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129671217</guid><author>m0_53371353</author><pubDate>Tue, 21 Mar 2023 12:06:45 +0800</pubDate><description><![CDATA[UniHPF：零领域知识的通用医疗预测框架]]></description><category></category></item><item><title><![CDATA[每天读一篇文献8--Contextualized Medication Information Extraction Using Transformer-based Deep Learning Arc]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129628083</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129628083</guid><author>m0_53371353</author><pubDate>Fri, 17 Mar 2023 22:44:14 +0800</pubDate><description><![CDATA[Contextualized Medication Information Extraction Using Transformer-based Deep Learning Architectures基于Transformer深度学习架构的情境化用药信息抽取]]></description><category></category></item><item><title><![CDATA[每天读一篇论文7--MEDBERT.de: A Comprehensive German BERT Model for the Medical Domain]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129603496</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129603496</guid><author>m0_53371353</author><pubDate>Thu, 16 Mar 2023 23:16:11 +0800</pubDate><description><![CDATA[MEDBERT.de: A Comprehensive German BERT Model for the Medical Domain一个全面的德国医学领域的BERT模型]]></description><category></category></item><item><title><![CDATA[每天读一篇论文6--Optimizing Deep Learning Model Parameters with the Bees Algorithm for Improved Medical Tex]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129559737</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129559737</guid><author>m0_53371353</author><pubDate>Wed, 15 Mar 2023 17:19:27 +0800</pubDate><description><![CDATA[Optimizing Deep Learning Model Parameters with the Bees Algorithm for Improved Medical Text Classification改进医学文本分类的Bees算法优化深度学习模型参数]]></description><category></category></item><item><title><![CDATA[每天读一篇论文5--MWE as WSD: Solving Multiword Expression Identification with Word Sense Disambiguation]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129538718</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129538718</guid><author>m0_53371353</author><pubDate>Tue, 14 Mar 2023 21:57:09 +0800</pubDate><description><![CDATA[MWE as WSD: Solving Multiword Expression Identification with Word Sense Disambiguation利用词义消歧解决多词表达识别问题]]></description><category></category></item><item><title><![CDATA[每天读一篇论文4]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129443577</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129443577</guid><author>m0_53371353</author><pubDate>Fri, 10 Mar 2023 15:21:58 +0800</pubDate><description><![CDATA[ICL-D3IE: In-Context Learning with Diverse Demonstrations Updating for Document Information Extraction面向文档信息抽取的多样演示更新上下文学习]]></description><category></category></item><item><title><![CDATA[每天读一篇论文3--Structure Pretraining and Prompt Tuning for Knowledge Graph Transfer]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129433443</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129433443</guid><author>m0_53371353</author><pubDate>Thu, 09 Mar 2023 23:12:18 +0800</pubDate><description><![CDATA[Structure Pretraining and Prompt Tuning for Knowledge Graph Transfer面向知识图谱迁移的结构预训练和快速调优]]></description><category></category></item><item><title><![CDATA[每天读一篇论文2--Document-level Relation Extraction with Cross-sentence Reasoning Graph]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129410622</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129410622</guid><author>m0_53371353</author><pubDate>Wed, 08 Mar 2023 21:54:20 +0800</pubDate><description><![CDATA[Document-level Relation Extraction with Cross-sentence Reasoning Graph基于跨句推理图的文档级关系抽取]]></description><category></category></item><item><title><![CDATA[每天读一篇论文1--ANCIENT CHINESE WORD SEGMENTATION AND PART-OF-SPEECH TAGGING USING DISTANT SUPERVISIO]]></title><link>https://blog.youkuaiyun.com/m0_53371353/article/details/129391756</link><guid>https://blog.youkuaiyun.com/m0_53371353/article/details/129391756</guid><author>m0_53371353</author><pubDate>Tue, 07 Mar 2023 23:27:44 +0800</pubDate><description><![CDATA[we propose a novel augmentation method of ancient Chinese WSG and POS tagging data using distant supervision over parallel corpus.我们提出了一种基于平行语料远程监督的古汉语WSG和词性标注数据扩充方法。]]></description><category></category></item></channel></rss>