检索结果-内蒙古大学图书馆

13th International conference on natural language processing and Chinese Computing

作者： Huang, Peijie Huang, Boxi Xu, Yuhong Chen, Weiting Li, Jia South China Agr Univ Coll Math & Informat Guangzhou Peoples R China

ISBN: (纸本)9789819794393;9789819794409

Utterance Domain Classification (UDC) is essential for Spoken language Understanding (SLU), a task analogous to short text classification. Short texts are often challenging to understand due to their lack of context, necessitating the enrichment of their semantic representation with supplementary information such as concepts from external knowledge bases. However, the inclusion of concepts introduces noise, making the selection of valuable concepts challenging. This paper proposes a UDC method employing keyword-guided signals to enhance the purity of external knowledge. We use two keyword extraction strategies to construct two types of keywords. A keyword-assisted concept denoising module addresses the concept noise problem, and a knowledge injection module is designed to better integrate concepts into the model. Experimental results on two Chinese SLU datasets demonstrate that our model achieves state-of-the-art performance.

关键词： Utterance domain classification Spoken language understanding Knowledge-enhanced methods

来源：评论

学校读者我要写书评

暂无评论

Mixed Multimodal Contrastive Learning for Enhancing Detection and Correction of Faked and Misspelled Chinese Characters 13th

Mixed Multimodal Contrastive Learning for Enhancing Detectio...

引用

13th International conference on natural language processing and Chinese Computing

作者： He, Rui Wang, Zhongqing Wang, Hongling Soochow Univ Nat Language Proc Lab Suzhou Peoples R China

ISBN: (纸本)9789819794423;9789819794430

Visual Chinese Character Checking (C3) aims to detect and correct errors in handwritten Chinese text images, including faked characters and misspelled characters. This task is beneficial for subsequent tasks by improving the efficiency of identifying errors in handwritten text. Recent methods are mainly based on Optical Character Recognition (OCR) and Pre-trained language Models (PLMs). Visual Chinese Character Checking is an emerging task, and relevant research has made progress. However, we believe that existing work has not fully leveraged the inherent knowledge of pre-trained models and has not addressed the semantic bias issue between pre-trained models and the character checking task. These challenges result in deficiencies in recognizing misspelled Chinese characters and correcting misused characters. Therefore, we propose various multimodal contrastive learning methods based on image-to-image and image-to-text comparisons. These methods are used throughout the processes of character recognition, error detection, and correction. By aligning the semantic feature representations among different models, our approach makes these models more suitable for the Visual Chinese Character Checking task, thereby enhancing their capabilities.

关键词： Chinese Character Checking Chinese Spell Checking Contrastive Learning Optical Character Recognition

来源：评论

学校读者我要写书评

暂无评论

A New Adaptive Prediction Algorithm for Judicial Sentencing with empirical Studies

引用

Journal of Systems Science & Complexity 2025年第1期38卷 3-20页

作者： DAI Ruifen WANG Fang GUO Lei ata Science Institute Shandong UniversityJinan250100China Key Laboratory of Systems and Control Academy of Mathematics and Systems ScienceChinese Academy of SciencesBeijing100190China

With the development and applications of the Smart Court System(SCS)in China,the reliability and accuracy of legal artificial intelligence have become focal points in recent ***,criminal sentencing prediction,a significant component of the SCS,has also garnered widespread *** to the Chinese criminal law,actual sentencing data exhibits a saturated property due to statutory penalty ranges,but this mechanism has been ignored by most existing *** this,the authors propose a sentencing prediction model that combines judicial sentencing mechanisms including saturated outputs and floating boundaries with neural *** on the saturated structure of our model,a more effective adaptive prediction algorithm will be constructed based on the fusion of several key ideas and techniques that include the utilization of the L1 loss together with the corresponding gradient update strategy,a data pre-processing method based on large language model to extract semantically complex sentencing elements using prior legal knowledge,the choice of appropriate initial conditions for the learning algorithm and the construction of a double-hidden-layer network *** empirical study on the crime of disguising or concealing proceeds of crime demonstrates that our method can achieve superior sentencing prediction accuracy and significantly outperform common baseline methods.

关键词： Adaptive prediction algorithm judicial mechanism neural networks sentencing prediction

来源：评论

学校读者我要写书评

暂无评论

Applying Text Mining to Analyze Human Question Asking in Creativity Research 12th

Applying Text Mining to Analyze Human Question Asking in C...

引用

12th International conference on Big Data Analytics, BDA 2024

作者： Wròblewska, Anna Korbin, Marceli Kenett, Yoed N. Dan, Daniel Ganzha, Maria Paprzycki, Marcin Faculty of Mathematics and Information Science Warsaw University of Technology Warsaw Poland Faculty of Data and Decision Sciences Technion - Israel Institute of Technology Haifa Israel School of Applied Data Science Modul University Vienna Vienna Austria Systems Research Institute Polish Academy of Sciences Warsaw Poland

ISBN: (纸本)9783031861925

Creativity relates to the ability to generate novel, and effective, ideas in the areas of interest. How are such creative ideas generated? One possible mechanism that supports creative ideation, and is gaining increased empirical attention, is by asking questions. Question asking is a likely cognitive mechanism that allows defining problems, facilitating creative problem solving [15]. However, much is unknown about the exact role of questions in creativity. This work presents an attempt to apply text mining methods to measure the cognitive potential of questions, taking into account, among others, (a) question type, (b) question complexity, and (c) content of the answer. This contribution summarizes the history of question mining as a part of creativity research, along with the natural language processing methods deemed useful, or helpful, in the study. Furthermore, a novel approach is proposed, implemented, and applied to five datasets. Obtained experimental results are comprehensively analyzed, suggesting that natural language processing has a role to play in the creativity research. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Creativity Measuring questions natural language processing Question asking

来源：评论

学校读者我要写书评

暂无评论

A Survey of Zero-Shot Stance Detection 13th

A Survey of Zero-Shot Stance Detection

引用

13th International conference on natural language processing and Chinese Computing

作者： Liu, Guangzhen Zhao, Kai Zhang, Linlin Bi, Xuehua Lv, Xiaoyi Chen, Cheng Xinjiang Univ Sch Comp Sci & Technol Urumqi 830049 Xinjiang Peoples R China Xinjiang Univ Sch Software Xi Urumqi 830049 Xinjiang Peoples R China Xinjiang Med Univ Coll Med Engn & Technol Urumqi 830017 Xinjiang Peoples R China Cent South Univ Sch Comp Sci & Engn Changsha 410083 Hunan Peoples R China

ISBN: (纸本)9789819794423;9789819794430

Stance detection is a subproblem of sentiment analysis, commonly defined as classifying the stance of a text toward a certain target: {Favor, Against, Neither}. As an important research problem, reliance on high-quality annotated data poses a significant challenge. However, in the real world, with the rapid development of social media, it is impossible to annotate the massive amount of text on diverse topics, a universal framework for stance detection is expected. Consequently, zero-shot stance detection methods that do not require annotated data have attracted researchers' attention. In this survey, we review and introduce the early research and definition of stance detection, with a focus on summarizing the current research status of ZSSD, discuss datasets and the most advanced models. Finally, based on the above research, we explore possible future directions.

关键词： natural language processing Deep learning Stance detection Zero-shot

来源：评论

学校读者我要写书评

暂无评论

Text pre-processing techniques in addressing fake news 4th

Text pre-processing techniques in addressing fake news

引用

4th International conference on Computational methods in Science and Technology, ICCMST 2024

作者： Dhiman, Pummy Kaur, Amandeep Chitkara University Institute of Engineering and Technology Chitkara University Punjab India

ISBN: (纸本)9781032911571

This study emphasizes the significance of identifying counterfeit news and the essentiality of proficient pre-processing approaches to clean text corpora. This work effectively integrated two Indian datasets and employed NLP techniques using Python programming language to carry out pre-processing activities, hence enhancing the data’s quality. The experiment part offers comprehensive insights into the procedures of combining and pre-processing. Although this study aids in the advancement of efficient pre-processing methods for identifying fake news, there remains a substantial amount of effort to enhance the precision of false news identification. In order to build a better-informed society and enhance our ability to detect false news, future research will investigate the use of DL approaches for the categorization of fake news. © 2025 the Author(s).

关键词： Dataset Fake news India natural language processing

来源：评论

学校读者我要写书评

暂无评论

Overview of the NLPCC2024 Shared Task 6: Scientific Literature Survey Generation 13th

Overview of the NLPCC2024 Shared Task 6: Scientific Literatu...

引用

13th International conference on natural language processing and Chinese Computing

作者： Tian, Yangjie Cul, Xungang Li, Aijia Zhang, He Xu, Ruohua Li, Yunfeng Liu, Ming Kexin Technol Beijing Peoples R China Deakin Univ Docklands Australia

ISBN: (纸本)9789819794423;9789819794430

With the relentless growth in the volume of academic publications and the accelerating speed of scholarly communication, the time researchers dedicate to literature surveys has become increasingly substantial. Automatic literature survey generation offers a valuable solution, liberating researchers from the time-intensive task of manually surveying the literature. We organized the NLPCC2024 Shared Task 6 for scientific literature survey generation. This paper will summarize the task information, the data set, the methods used by participants and the final results. Furthermore, we will discuss key findings and challenges for scientific literature survey generation in the scientific domain.

关键词： Scientific Literature Survey Generation Large language Model Generative Pre-trained Transformer (GPT) In-context Learning Retrieval-Augmented Generation (RAG)

来源：评论

学校读者我要写书评

暂无评论

Detoxifying Large language Models via Kahneman-Tversky Optimization 13th

Detoxifying Large Language Models via Kahneman-Tversky Optim...

引用

13th International conference on natural language processing and Chinese Computing

作者： Li, Qingquan Du, Wenlong Liu, Jin Ant Grp Hangzhou Peoples R China

ISBN: (纸本)9789819794423;9789819794430

Currently, the application of Large language Models (LLMs) faces significant security threats. Harmful questions and adversarial attack prompts can induce the LLMs to generate toxic responses. Therefore, detoxifying LLMs is a critical research topic to ensure their safe and widespread application. In this paper, we propose an alignment-based detoxification method for LLMs. We utilize Kahneman-Tversky Optimization (KTO) to align LLMs. During the construction of the training dataset, we take into account both the detoxification performance and the potential side effect on the LLMs. For detoxification, we make the LLM preferentially generate safe responses rather than toxic contents when asked with harmful questions and attack prompts. To mitigate the potential side effect on the conversational capabilities of LLMs, we incorporate normal questions into the training data, and ensure that the LLM generate normal answers, rather than safety refusals or unsafe responses. Experimental results show that our method showcase the best detoxification performance among all baseline methods while exerting little negative impact on the LLMs. Moreover, our method even enhance the LLMs' general abilities such as question answering and language understanding. Our proposed method achieve the first place in the NLPCC 2024 Share Task 10 Track 2 with an average score of 52.31.

关键词： Large language models Detoxification Alignment

来源：评论

学校读者我要写书评

暂无评论

Leveraging Large language Models Knowledge Enhancement Dual-Stage Fine-Tuning Framework for Recommendation 13th

Leveraging Large Language Models Knowledge Enhancement Dual-...

引用

13th International conference on natural language processing and Chinese Computing

作者： Zeng, Biqing Shi, Hao Li, Yangyu Li, Ruizhe Deng, Huimin South China Normal Univ Sch Software Foshan Peoples R China South China Normal Univ Aberdeen Inst Data Sci & Artificial Intelligence Foshan Peoples R China Univ Aberdeen Dept Comp Sci Aberdeen Scotland Guangdong AIB Polytech Sch Comp Sci Guangzhou Peoples R China

ISBN: (纸本)9789819794331;9789819794348

Large language models(LLMs) have exhibited notable general-purpose task-solving abilities in language understanding and generation, including processing recommendation tasks. The majority of existing research relies on training-free recommendation models that treat LLMs as reasoning engines and directly given the recommended task response. This approach heavily relies on pre-trained knowledge and may lead to excessive costs. As such, we propose a two-stage fine-tuning framework leveraging LLaMA2 and GPT-4 Knowledge Enhancement for recommendation. In particular, we use GPT-4 Instruction-Following data to tune the LLM in first-stage instruction tuning process, achieving lower training costs and better inference performance. In the second stage, through a elaborately designed prompt template, we fine-tune LLM from the first stage in a few-shot setting by interactive sequences based on user ratings. To validate the effectiveness of our framework, we compare against state-of-the-art baseline methods on benchmark datasets. The results demonstrate that our framework has promising recommendation capabilities. Our experiments are executed on a single RTX4090 with LLaMA2-7B.

关键词： Large language models Knowledge Enhancement Recommendation Instruction Tuning

来源：评论

学校读者我要写书评

暂无评论

Federated Prototype Guided Adaption for Vision-language Models

Federated Prototype Guided Adaption for Vision-Language Mode...

引用

2025 IEEE International conference on Acoustics, Speech, and Signal processing, ICASSP 2025

作者： Liu, Youchao Huang, Dingjiang School of Data Science and Engineering East China Normal University Shanghai China

ISBN: (纸本)9798350368741

Federated Learning (FL) is a new pivotal paradigm for decentralized training on heterogeneous data. Recently fine-tuning of Vision-language Models (VLMs) has been extended to the federated setting to improve overall performance. Unfortunately, in this case, FL still faces two critical challenges that hinder its actual performance: data distribution heterogeneity and high resource costs brought by large VLMs. In this paper, we introduce FedPGA, a prototype-guided method, for achieving performance improvements in the federated setting for VLMs. Concretely, we design a prototype-based adapter for the vision-language model, CLIP. The lightweight adapter updates the prior knowledge encoded in CLIP to enhance its adaption capability further and avoid the effects of data distribution heterogeneity in the federated setting. Simultaneously, small-scale operations can mitigate the computational and communication burden caused by large VLMs. Our comprehensive empirical evaluations of nine diverse image classification datasets show that our method is superior to existing FL methods under VLMs. © 2025 IEEE.

关键词： federated learning few-shot classification prototype-based adapter vision-language model

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：