检索结果-内蒙古大学图书馆

AT-FinGPT: Financial risk prediction via an audio-text large language model

FINANCE RESEARCH LETTERS 2025年 77卷

作者： Liu, Yingnan Bu, Ningbo Li, Zhiqiang Zhang, Yongmin Zhao, Zhenyu Ningbo Univ Sch Business Ningbo 315211 Peoples R China Chinese Acad Sci Ningbo Inst Mat Technol & Engn NlMTE Lab Robot & Intelligent Mfg Equipment Technol Ningbo 315201 Peoples R China Donghai Lab Zhoushan 316021 Peoples R China Ningbo Univ Sch Law Ningbo 315211 Peoples R China

Financial risk prediction is crucial for investment decision-making. Traditional machine learning methods are limited by their structures and parameter size, which hinders their generalizability and effectiveness. Large language models (LLMs), which are pretrained with very large dataset and many GPUs have recently shown promising improvements in financial risk prediction. Despite this progress, most existing financial LLMs mainly rely on textual data for training and prediction, overlooking audio data and limiting analysis to text summarization. However, natural language processing studies have shown that audio from CEOs' quarterly earnings calls is crucial for financial risk prediction. In this work, we introduce an audio-text LLM named ATFinGPT, which fuses financial audio data and summarization texts for financial risk prediction. The empirical experimental results show that AT-FinGPT is superior to most advanced methods. Through an ablation study, we demonstrate that different data sources can facilitate financial risk assessment and discuss the effectiveness of each part in the AT-FinGPT model.

关键词： Quantitative finance Large language model FinGPT Multi-sources data fusion

来源：评论

学校读者我要写书评

暂无评论

Automatic Text Simplification for the Legal Domain in Brazilian Portuguese 34th

Automatic Text Simplification for the Legal Domain in Brazil...

引用

34th Brazilian conference on Intelligent Systems

作者： Pereira, Francielle Vasconcellos Frazao, Ana Moreira, Viviane P. UFRGS Porto Alegre Inst Informat Porto Alegre RS Brazil USP Sao Paulo Sao Paulo Brazil

ISBN: (纸本)9783031790379;9783031790386

Legal and juridical documents such as rulings, laws, agreements, and contracts contain domain-specific terms and jargon, long and complex sentences that may be difficult to understand for laypeople without domain expertise, reading issues, or with a low education level. The simplification of these documents has been a concern for several years, aiming to democratize access to justice. Courts are already adopting simpler language, especially in documents aimed at laypeople, such as warrants and notifications, to enhance inclusion and clarity. Automatic textual simplification, a subfield of natural language processing, seeks to make complex texts more accessible. This paper explores the task of automatic text simplification in Portuguese for the legal domain. The main challenge here is the lack of datasets containing complex sentences and their simplified versions. This work investigates how existing datasets, methods, and metrics used for text simplification perform applied to legal texts in Portuguese. We present qualitative and quantitative analyses using five models. The results show that GPT-based models have the best results, but fine-tuning with domain data is a viable open-source alternative.

关键词： Automatic Text Simplification Legal Texts natural language processing Plain language

来源：评论

学校读者我要写书评

暂无评论

MLEE: Event Extraction as Multi-label Classification Task at Token Level 6th

MLEE: Event Extraction as Multi-label Classification Task a...

引用

6th IFIP TC 12 International conference on Intelligence Science, ICIS 2024

作者： Yang, Jinshun Huang, Shuangxi Huang, Mingfeng Tsinghua University Beijing China

ISBN: (纸本)9783031712524

Event Extraction is an important task in natural language understanding, which aims to identify event trigger of pre-defined event types and their arguments of specific roles, has attracted a lot of attention from industry and academia. The previous works failed to address some issues, including error propagation problem, overlap and nest problem, and high complexity of model. This work proposes a novel model MLEE, which models Event Extraction task as Multi-Label classification task at token level, and processes the extraction task in a joint paradigm, can help solving issues mentioned above. The experiment verifies our model’s effectiveness. empirical results on DuEE and FewFC shows that MLEE outperforms previous best model, pushing trigger extraction F1 to 85.03% (+4.45%), argument extraction F1 to 78.85% (+2.72%) on DuEE, pushing trigger extraction F1 to 76.69% (+1.63%), argument extraction F1 to 76.53% (+5.27%) on FewFC. © IFIP International Federation for Information processing 2025.

关键词： natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Retrieval-Augmented Code Generation for Universal Information Extraction 13th

Retrieval-Augmented Code Generation for Universal Informatio...

引用

13th International conference on natural language processing and Chinese Computing

作者： Guo, Yucan Li, Zixuan Jin, Xiaolong Liu, Yantao Zeng, Yutao Liu, Wenxuan Li, Xiang Yang, Pan Bai, Long Guo, Jiafeng Chen, Xueqi Chinese Acad Sci Inst Comp Technol CAS Key Lab Network Data Sci & Technol Beijing Peoples R China Univ Chinese Acad Sci Sch Comp Sci & Technol Beijing Peoples R China

ISBN: (纸本)9789819794331;9789819794348

Information Extraction (IE) aims to extract structural knowledge (e.g., entities, relations, events) from natural language texts. Recently, Large language Models (LLMs) with code-style prompts have demonstrated powerful capabilities in IE tasks. However, adopting code LLMs to conduct IE tasks still has two challenges: (1) It still lacks a unified code-style prompt for different IE tasks since existing methods use task-specific prompts for separate IE tasks. (2) It still lacks an effective in-context learning (ICL) method to encourage LLMs to conduct IE tasks precisely, considering some powerful LLMs are close-sourced and not trainable. Therefore, this paper proposes a code generation framework for Universal IE (UIE) tasks called Code4UIE. Specifically, for the first challenge, Code4UIE designs a unified code-style schema for various IE tasks via Python classes. By so doing, different IE tasks can be associated, and LLMs can learn from various IE tasks effectively. For the second challenge, Code4UIE adopts a retrieval-augmented mechanism to comprehensively utilize the ICL ability of LLMs. Extensive experiments on five representative IE tasks across nine datasets demonstrate the effectiveness of the Code4UIE framework.

关键词： Information Extraction Code Generation Large language Model Information Retrieval

来源：评论

学校读者我要写书评

暂无评论

Multilingual Extractive Summarization: Investigating State-of-the-Art methods for English and Brazilian Portuguese 34th

Multilingual Extractive Summarization: Investigating State-o...

引用

34th Brazilian conference on Intelligent Systems

作者： Jorge, Germano Antonio Zani Bezerra, Davi Alves Xavier, Clarissa Castella Pardo, Thiago Alexandre Salgueiro Univ Sao Paulo Nucleo Interinstituc Linguist Computac NILC Inst Cencias Matemat & Computacao Sao Carlos SP Brazil SiDi Campinas SP Brazil

ISBN: (纸本)9783031790317;9783031790324

Automatic Text Summarization (ATS) is a natural language processing (NLP) task essential for handling large volumes of information. ATS can be classified into two main types: extractive and abstractive. Extractive summarization selects sentences or phrases directly from the source text(s), while abstractive summarization generates new sentences that try to capture the original meaning of the source text(s). This paper describes our efforts to perform extractive single-document summarization in multilingual contexts. Although various summarization methods, such as PreSumm and HiStruct+, have shown promising results on English corpora like CNN/DM, there is a significant gap in applying these methods to other languages, especially Brazilian Portuguese. Additionally, these summarizers were evaluated with traditional metrics like ROUGE, which has limitations as it primarily measures superficial text overlap. To fill these gaps, we evaluate the effectiveness of these state-of-the-art methods on the CSTNews corpus (with news texts in Brazilian Portuguese) employing ROUGE and the recent BLANC metric, which measures how much the generated summary aids a pre-trained language model (like BERT) in understanding the document. Our contributions include the results and comparison of adapted models, the discussion of the BLANC metric in contrast to ROUGE, and the expansion of resources available to the Portuguese and multilingual NLP community.

关键词： Extractive Summarization Multilinguality natural language processing

来源：评论

学校读者我要写书评

暂无评论

WireLlama - Agentic Query System for PCAP Analysis 12th

WireLlama - Agentic Query System for PCAP Analysis

引用

12th International conference on Big Data Analytics, BDA 2024

作者： Choudhary, Milind Kaur, Parmeet Jaypee Institute of Information Technology Noida India

ISBN: (纸本)9783031861925

Network security researchers and practitioners constantly seek improved methods to manage and interpret vast amounts of network log data. The paper presents WireLlama, an Agentic Query system tailored for packet capture (PCAP) analysis. By leveraging the llama-agents and Scapy frameworks, WireLlama aims to enhance the capability of deriving insights from network traffic data. A pre-trained LLaMA model is utilized from Hugging Face for processing of user queries in natural language. These are translated into SQL and then the model then generates responses based on the SQL query and the preprocessed network data. The responses are also provided in natural language for easy comprehension by the user. As illustrated by results, the proposed system is successful in handling specific query functions, like identifying top IP addresses by traffic volume or detecting any suspicious activity. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Structured Query language

来源：评论

学校读者我要写书评

暂无评论

Semantic Knowledge Enhanced and Global Pointer Optimized Method for Medical Nested Entity Recognition 13th

Semantic Knowledge Enhanced and Global Pointer Optimized Met...

引用

13th International conference on natural language processing and Chinese Computing

作者： Song, Yilin Kong, Fang Soochow Univ Lab Nat Language Proc Suzhou Peoples R China

ISBN: (纸本)9789819794300;9789819794317

In the medical field, unstructured medical text holds rich medical knowledge. Identifying medical entities in this text accurately is crucial for structured medical databases, knowledge graphs, and intelligent diagnostic systems. Medical text has unique features, making it hard for traditional NER methods to identify complex medical entities. In particular, the recognition of nested entities within medical text poses a significant challenge, as it requires systems to recognize and understand the complex hierarchical relationships between entities, placing higher demands on traditional entity recognition systems. To overcome the challenges of nested entity recognition in medical text, we propose a method that combines semantic knowledge enhancement and global pointer optimization. Initially, we incorporate semantic prior knowledge of entity categories, capturing the interplay between labels and text by integrating label relationships. This allows us to obtain candidate entity information enriched with integrated label details. Following this, we establish a classification module to evaluate and score these candidate entities along with their labels, enabling entity prediction. To address nested entities, we introduce a Efficient GlobalPointer module that computes the likelihood of each text span being a specific entity type, thus bolstering nested entity recognition. By merging the outputs from both modules, we arrive at the final predicted entities. Experimental results indicate that our method excels on two flat entity datasets, CMedQANER and CCKS2017, as well as on the nested entity dataset CMeEE. Compared to baseline models, our approach demonstrates notable performance enhancements.

关键词： Named Entity Recognition Nested Entity Recognition Medical Named Entity Recognition

来源：评论

学校读者我要写书评

暂无评论

Key Algorithms for Keyphrase Generation: Instruction-Based LLMs for Russian Scientific Keyphrases 12th

Key Algorithms for Keyphrase Generation: Instruction-Based ...

引用

12th International conference on Analysis of Images, Social Networks and Texts, AIST 2024

作者： Glazkova, Anna Morozov, Dmitry Garipov, Timur University of Tyumen Tyumen Russia Novosibirsk State University Novosibirsk Russia

ISBN: (纸本)9783031880353

Keyphrase selection is a challenging task in natural language processing that has a wide range of applications. Adapting existing supervised and unsupervised solutions for the Russian language faces several limitations due to the rich morphology of Russian and the limited number of training datasets available. Recent studies conducted on English texts show that large language models (LLMs) successfully address the task of generating keyphrases. LLMs allow achieving impressive results without task-specific fine-tuning, using text prompts instead. In this work, we access the performance of prompt-based methods for generating keyphrases for Russian scientific abstracts. First, we compare the performance of zero-shot and few-shot prompt-based methods, fine-tuned models, and unsupervised methods. Then we assess strategies for selecting keyphrase examples in a few-shot setting. We present the outcomes of human evaluation of the generated keyphrases and analyze the strengths and weaknesses of the models through expert assessment. Our results suggest that prompt-based methods can outperform common baselines even using simple text prompts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Automated Framework for Constructing Knowledge Graphs Oriented for Standard Analysis Using Large language Models 24

Automated Framework for Constructing Knowledge Graphs Orient...

引用

2nd International conference on Electronics, Computers and Communication Technology, CECCT 2024

作者： Zhu, Sitong Xia, Baobing Duan, Fangwei Zhao, Zhenyang Zhao, Zhenxia Xiao, Teliang Liu, Zhicheng Liu, Xia State Grid Liaoning Electric Power Co. LTD. Electric power Research Institute Liaoning Shenyang China Beijing State Grid Communication Accenture Information Technology Co. Ltd. Beijing China

ISBN: (纸本)9798400710193

In an era characterized by rapid technological growth and digital transformation, the necessity for efficient and structured knowledge representation has grown more crucial. Standards serve as fundamental cornerstones that offer guidelines, specifications, and frameworks to ensure the quality and interoperability of products, services, as well as systems. Nonetheless, the complexity and extensive nature of standard documents present significant challenges in extraction, alignment, and organization. Traditional manual processing methods frequently prove labor-intensive and susceptible to errors, impeding the capturing of intricate relationships and hierarchies within these standards. Knowledge Graphs (KGs) provide a robust methodology for organizing information, facilitating improved functionalities for advanced search, reasoning, and analytics. Despite their potential, structuring KGs from standard documents continues to be challenging because of unstructured text, domain-specific terminology, and the intricacies in accurate information extraction. Recent advancements in natural language processing (NLP), particularly the emergence of Large language Models (LLMs) like GPT-3, have opened new avenues for automating the extraction and structuring of information from unstructured content. These models exhibit exceptional proficiency in comprehending and producing human-like text, positioning them as feasible solutions for addressing the complexities associated with standard documents. This paper presents an automated framework named StandardKG Builder, which utilizes LLMs to construct knowledge graphs tailored for standard analysis from multiple perspectives for complex information extraction. Our evaluation on a comprehensive dataset of standard documents highlights the framework’s effectiveness and scalability. By merging sophisticated knowledge representation with advanced NLP techniques, our work significantly enhances the accessibility and analysis of standard docu

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

LTNER: Large language Model Tagging for Named Entity Recognition with Contextualized Entity Marking 27th

LTNER: Large Language Model Tagging for Named Entity Recogni...

引用

27th International conference on Pattern Recognition, ICPR 2024

作者： Yan, Faren Yu, Peng Chen, Xin Computer Network Information Center Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9783031784941

The use of LLMs for natural language processing has become a popular trend in the past two years, driven by their formidable capacity for context comprehension and learning, which has inspired a wave of research from academics and industry professionals. However, for certain NLP tasks, such as NER, the performance of LLMs still falls short when compared to supervised learning methods. In our research, we developed a NER processing framework called LTNER (Code is available at https://***/YFR718/LTNER) that incorporates a revolutionary Contextualized Entity Marking Gen Method. By leveraging the cost-effective GPT-3.5 coupled with context learning that does not require additional training, we significantly improved the accuracy of LLMs in handling NER tasks. The F1 score on the CoNLL03 dataset increased from the initial 85.9% to 91.9%, approaching the performance of supervised fine-tuning. This outcome has led to a deeper understanding of the potential of LLMs. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： natural language processing systems

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：