检索结果-内蒙古大学图书馆

35th international conference on Software engineering and knowledge engineering, SEKE 2023

作者： Xie, Zuotong Ouyang, Kai Xie, Xiangjin Zheng, Hai-Tao Liu, Wenqiang Huang, Dongxiao Wu, Bei Shenzhen International Graduate School Tsinghua University China Interactive Entertainment Group Tencent Inc China Pengcheng Laboratory Shenzhen China

Pre-trained language models (PLMs) with external knowledge have demonstrated their remarkable performance on a variety of downstream natural language processing tasks. The typical methods of integrating knowledge into PLMs are designing different pre-training tasks and training from scratch, which requires high-end hardware, massive storage resources, and computing time. Prompt learning is an effective approach to tune PLMs for specific tasks, and it can also be used to infuse knowledge. However, most prompt learning methods accept one token as the answer instead of multiple tokens. To tackle this problem, we propose the long-answer prompt learning method (KLAPrompt) to incorporate semantic knowledge in Xinhua Dictionary into pre-trained language models. The proposed method splits the whole answer space into several answer subspaces according to the token’s position in the long answer. Extensive experimental results on five datasets demonstrate the effectiveness of our approach. © 2023 knowledge Systems Institute Graduate School. All rights reserved.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

The Research and Construction of the AI-Based knowledge Graph in Multi-Dimensional Data

The Research and Construction of the AI-Based Knowledge Grap...

引用

2023 international conference on Computer engineering and Distance Learning, CEDL 2023

作者： Zhou, Jing Wuhan Business University School of Information Engineering Hubei Province Wuhan City China

ISBN: (纸本)9798350329032

With the development of deep learning and the knowledge graph, artificial intelligence has had a significant influence on the area of education as a result of the rapid growth of this age, which has profoundly altered human productivity and daily life. The university is undergoing a digital transformation. The essence of the knowledge graph is the knowledge base of the semantic network. Using natural language processing (NLP) technology, the university can construct a knowledge graph. Integrating knowledge graphs and deep learning has become one of the most important aspects of further improving the effect of deep learning. The solution to knowledge graph processing data is 'algorithm, totalization, and implementation'. The key technologies of constructing knowledge graphs in multi-dimensional data mainly focus on knowledge ontology definition, knowledge representation, knowledge modeling, knowledge extraction, knowledge fusion, knowledge processing, knowledge computing, and other technologies. The research in this paper shows that the RDF model and algorithm of the knowledge graph have application prospects in multi-dimensional data. © 2023 IEEE.

关键词： knowledge graph

来源：评论

学校读者我要写书评

暂无评论

FinMoE: A MoE-based Large Chinese Financial language Model 9

FinMoE: A MoE-based Large Chinese Financial Language Model

引用

Joint Workshop of the 9th Financial Technology and natural language processing, FinNLP 2025, the 6th Financial Narrative processing, FNP 2025, and the 1st Workshop on Large language Models for Finance and Legal, LLMFinLegal 2025, co-located with the 31st international conference on Computational Linguistics, COLING 2025

作者： Zhang, Xuanyu Yang, Qing Du Xiaoman Financial Beijing China

ISBN: (纸本)9798891762091

Large-scale language models have demonstrated remarkable success, achieving strong performance across a variety of general tasks. However, when applied to domain-specific fields, such as finance, these models face challenges due to the need for both specialized knowledge and robust general capabilities. In this paper, we introduce FinMoE, a MOE-based large-scale Chinese financial language model that bridges the gap between general language models and domain-specific requirements. FinMoE employs a dense MoE architecture, where all expert networks are simultaneously activated and dynamically combined to effectively integrate general linguistic understanding with domain-specific financial expertise. Experimental results demonstrate that FinMoE achieves state-of-the-art performance on both general-purpose and financial benchmarks at a comparable scale, validating its ability to balance domain specialization with general knowledge and reasoning. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Attention Distillation for Accuracy Improvement of Vision Transformer 12

Attention Distillation for Accuracy Improvement of Vision Tr...

引用

12th IEEE international conference on Identification, Information and knowledge in the Internet of Things, IIKI 2024

作者： Tanaka, Taiga Ishibashi, Ryuto Meng, Lin Ritsumeikan University Graduate School of Science and Engineering 1-1-1 Noji-higashi Shiga Kusatsu525-8577 Japan Ritsumeikan University College of Science and Engineering 1-1-1 Noji-higashi Shiga Kusatsu525-8577 Japan

ISBN: (纸本)9798331510633

In recent years, rapid advancements in technology have propelled progress in the fields of image recognition and natural language processing, resulting in the development of numerous applications and services. In particular, the advancement of deep learning has greatly contributed to improving the accuracy of these fields, and the use of AI is becoming more widespread in everyday life [1]. Among them, Vision Transformers (ViT) has emerged as a robust architecture in image recognition. ViT requires a large dataset for training to improve accuracy. To solve this problem, DeiT using knowledge distillation has been proposed. However, this DeiT does not consider each token's contribution to [CLS] in learning. In this study, a knowledge distillation model training method called ADViT that uses the contribution of each token to Classify token ([CLS]) is proposed. Specifically, the attention map of each token for [CLS] is calculated. In each layer, learning is performed to bring the student model's attention map closer to the teacher model's attention map. Then, the accuracy is evaluated, and experimental results show the effectiveness of the proposed method, with accuracy rates of ADViT-Tiny(L1): 98.35%, ADViT-Tiny(L2): 98.06%, ADViT-Small(L1):99.00%, and ADViT-Small(L2): 98.75%, which are higher than those of DeiT, DeiT-Tiny: 96.39%, DeiT-Small: 96.83%. Future work includes managing the number of heads and reducing the computation for more optimization. © 2024 IEEE.

关键词： natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

language Models over Large-Scale knowledge Base: on Capacity, Flexibility and Reasoning for New Facts 31

Language Models over Large-Scale Knowledge Base: on Capacity...

引用

31st international conference on Computational Linguistics, COLING 2025

作者： He, Qiyuan Wang, Yizhong Yu, Jianfei Wang, Wenya College of Computing and Data Science Nanyang Technological University Singapore Paul G. Allen School of Computer Science & Engineering University of Washington United States School of Computer Science and Engineering Nanjing University of Science and Technology China

ISBN: (纸本)9798891761964

Advancements in language models (LMs) have sparked interest in exploring their potential as knowledge bases (KBs) due to their high capability for storing huge amounts of factual knowledge and semantic understanding. However, existing studies face challenges in quantifying the extent of large-scale knowledge packed into LMs and lack systematic studies on LMs' structured reasoning capabilities over the infused knowledge. Addressing these gaps, our research investigates whether LMs can effectively act as large-scale KBs after training over an expansive set of world knowledge triplets via addressing the following three crucial questions: (1) How do LMs of different sizes perform at storing world knowledge of different frequencies in a large-scale KB? (2) How flexible are these LMs in recalling the stored knowledge when prompted with natural language queries? (3) After training on the abundant world knowledge, can LMs additionally gain the ability to reason over such information to infer new facts? Our findings indicate that while medium-scaled LMs hold promise as world knowledge bases capable of storing and responding with flexibility, enhancements in their reasoning capabilities are necessary to fully realize their potential. © 2025 Association for Computational Linguistics.

关键词： Structured Query language

来源：评论

学校读者我要写书评

暂无评论

Role of natural language processing and Machine Learning on (MTS)-Machine Translation System for Indian languages: A Review 15

Role of Natural Language Processing and Machine Learning on ...

引用

15th international conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Megha, V. Pooja, M.R. Vidyavardhaka College of Engineering Mysuru India

ISBN: (纸本)9798331300579

The conversion process of one specific language to another either in a completely automatic manner or with considerable amount of human intervention by preserving the original meaning of the input source text is known as (MTS) Machine Translation System. The expeditious availability of data and the need for increased processing speed of information has led the way towards natural language processing, Machine Learning, Speech processing, Image processing etc. When it comes to the Indian regional languages the research in this domain is very minimal as an outcome there is limited resources. In this paper, the researcher will be able to know the general workflow of how the translation process of source to destination language will occur and the various approaches and the methods for processing Indian languages proposed by scientists which are verified and validated, the challenges a researcher would face while developing one such translation tool followed by the summary of the entire review. © Grenze Scientific Society, 2024.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Simulation of Structured natural language Algorithm on Cloud Computing 3

Simulation of Structured Natural Language Algorithm on Cloud...

引用

3rd IEEE international conference on Electrical engineering, Big Data and Algorithms, EEBDA 2024

作者： Fang, Quanrong Tang, Keyin Investment Co. Ltd. Beijing China

ISBN: (纸本)9798350380989

From the sentiment analysis of social media to the monitoring of public opinion in the financial field, the application of natural language analysis continues to expand, and people's demand for natural language processing technology is also increasing. However, traditional natural language analysis algorithms face performance bottlenecks and energy consumption. Therefore, this paper proposes a structured natural language analysis algorithm based on cloud computing. Through simulation experiments, the performance and energy consumption of the algorithm are tested and evaluated. After a comprehensive experimental scheme design, the experimental group is compared with the control group by selecting the appropriate data set and establishing appropriate evaluation indicators, while configuring the experimental environment. The experimental results show that the lowest energy consumption of the algorithm in this paper is 2.41%, and the structured natural language analysis algorithm based on cloud computing has obvious advantages over traditional methods. The proportion of energy consumption in the experimental group is lower than that of the control group, indicating that the algorithm of the experimental group is better in terms of energy utilization efficiency. The performance degradation of the experimental group is lower when processing large-scale data, indicating that the experimental group algorithm has better extensibility and efficiency. These results provide strong support for the application of structured natural language analysis algorithms based on cloud computing, which helps to improve system efficiency and sustainability. © 2024 IEEE.

关键词： algorithm simulation cloud computing natural language structured analysis

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis Using LSTM Model for Emotion Detection 8

Sentiment Analysis Using LSTM Model for Emotion Detection

引用

8th international conference on Electronics, Materials engineering and Nano-Technology, IEMENTech 2025

作者： Dam, Mehuli Roy, Sumanta Sarkar, Darothi Mahata, Sainik Kumar Computer Science and Engineering Institute of Engineering & Management Kolkata University of Engineering and Management Kolkata India

ISBN: (纸本)9798331520762

This research focuses on the role of natural language processing in extracting, analysing, and interpreting textual data to identify sentiment and emotional patterns from tweets. NLP techniques, particularly deep learning techniques such as LSTM, allow for processing large-scale unstructured data while capturing contextual relationships within the text. The results of this study contribute to the NLP analysis and a wider area of mood, and understand the emotional tone of politicians and the perception of the general public. These results emphasize the importance of deep training models that have advanced in the analysis of political discourse, and this work is a social network researcher that tries to control and understand public mood in real time. In this study, we investigate the use of long -term data in memory networks to analyse the mood of the US President Donald Trump and Joe Biden. The study uses a dataset of over one million tweets and employs advanced natural language processing techniques such as text pre-processing, tokenization, and word embeddings to classify sentiment into positive, negative, or neutral categories. LSTM has the ability to capture long-term dependencies in sequential data and effectively resolves the complexities of informal, context-rich language in social networks. The proposed model achieved good accuracy and demonstrated its robustness in handling large-scale unstructured data. © 2025 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Converse Attention knowledge Transfer for Low-Resource Named Entity Recognition

引用

international Journal of Crowd Science 2024年第3期8卷 140-148页

作者： Lyu, Shengfei Sun, Linghao Yi, Huixiong Liu, Yong Chen, Huanhuan Miao, Chunyan School of Computer Science and Technology University of Science and Technology of China Hefei230027 China School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore

In recent years, great success has been achieved in many tasks of natural language processing (NLP), e.g., named entity recognition (NER), especially in the high-resource language, i.e., English, thanks in part to the considerable amount of labeled resources. More labeled resources, better word representations. However, most low-resource languages do not have such an abundance of labeled data as high-resource English, leading to poor performance of NER in these low-resource languages due to poor word representations. In the paper, we propose converse attention network (CAN) to augment word representations in low-resource languages from the high-resource language, improving the performance of NER in low-resource languages by transferring knowledge learned in the high-resource language. CAN first translates sentences in low-resource languages into high-resource English using an attention-based translation module. In the process of translation, CAN obtains the attention matrices that align word representations of high-resource language space and low-resource language space. Furthermore, CAN augments word representations learned in low-resource language space with word representations learned in high-resource language space using the attention matrices. Experiments on four low-resource NER datasets show that CAN achieves consistent and significant performance improvements, which indicates the effectiveness of CAN. © The author(s) 2024.

关键词： Labeled data

来源：评论

学校读者我要写书评

暂无评论

Cross-language Code Matching with IR2Vec and Cross-Entropy Loss 24

Cross-Language Code Matching with IR2Vec and Cross-Entropy L...

引用

8th international conference on natural language processing and Information Retrieval, NLPIR 2024

作者： Ghosh, Sagarika Das, Soma Chatterji, Sanjay Pratihar, Sanjoy CSE Indian Institute of Information Technology Kalyani West Bengal Kalyani India CSE University of Engineering and Management Jaipur Rajasthan Jaipur India CSE Institute of Engineering and Management Kolkata West Bengal Kolkata India

ISBN: (纸本)9798400717383

In contemporary software development and maintenance, the practice of reusing or copy-pasting code is prevalent to expedite processes. Consequently, numerous techniques have been developed to detect, compare, or match similar code within the same programming language. However, there remains a need for methods capable of detecting, comparing, or matching similar code across different programming languages. This paper presents a neural model utilizing IR2Vec embeddings and cross-entropy loss to facilitate the matching of similar code between disparate languages, such as C and C++. The proposed model achieves a matching accuracy of . To evaluate the model's efficacy, we compute Precision, Recall, and F1 score across various threshold values. © 2024 Copyright held by the owner/author(s).

关键词： Software design

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：