检索结果-内蒙古大学图书馆

12th National CCF Conference on Natural Language Processing and Chinese computing, NLPCC 2023

作者： Wang, Huijie He, Ruifang Jia, Yungang Xu, Jing Wang, Bo College of Intelligence and Computing Tianjin University Tianjin China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin China Tianjin Branch of National Computer Network and Information Security Management Center Tianjin China

ISBN: (纸本)9783031446924

Multi-turn dialogue response generation aims to generate a response with consideration of the context. It is not equal to multiple single-turn dialogues due to the context dependence of response. Many existing models achieve great success for response generation, but they still struggle to model the contextual semantics of dialogue history. Sequence models have difficulties to explore the interactive relations between contextual utterances, which affects the coherence of generated responses. To solve the issue, we propose a discourse relation-aware model, which encodes the contextual utterances with a directed acyclic graph neural network (DAGNN) with constraints on dialogue-specific discourse relations to better model the intrinsic structure. Besides, we introduce an auxiliary discourse relation recognition task to enhance the model’s ability of representing the context. Extensive experimental results show that our proposed model outperforms baselines. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Directed graphs

来源：评论

学校读者我要写书评

暂无评论

Dual-Prompting Interaction with Entity Representation Enhancement for Event Argument Extraction 12th

Dual-Prompting Interaction with Entity Representation Enhan...

引用

12th National CCF Conference on Natural Language Processing and Chinese computing, NLPCC 2023

作者： He, Ruifang Xiao, Mengnan Ma, Jinsong Zhang, Junwei Zhao, Haodong Zhang, Shiqi Bai, Jie College of Intelligence and Computing Tianjin University Tianjin300350 China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin300350 China The 54th Research Institute of CETC Shijiazhuang050081 China

ISBN: (纸本)9783031446955

Event argument extraction (EAE) aims to recognize arguments that are entities involved in events and their roles. Previous prompt-based methods focus on designing appropriate prompt template for events, while neglecting to explicitly model entities in the input, resulting in models that are not well aware of arguments. In this paper, we propose a novel Dual-Prompting Interaction approach with Entity Representation Enhancement (DPIERE) to explicitly model entities for both sentence-level and document-level EAE. 1) We design appropriate event and input prompt templates to model argument roles and entities in the input respectively so that pre-trained language model (PLM) can better capture the interaction between them. 2) We introduce a simple but effective entity representation enhancement method with adaptive selection to infuse entity knowledge into PLM. Finally, Argument span prediction is employed to predict the start and end word among the input for each role. In addition, DPIERE devises position marks in event prompt template to distinguish multiple occurrences of the same argument role. Comprehensive experimental results on three benchmarks show the effectiveness of our proposed approach. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

SELF-SUPERVISED AUDIO-VISUAL SPEAKER REPRESENTATION WITH CO-META LEARNING 48

SELF-SUPERVISED AUDIO-VISUAL SPEAKER REPRESENTATION WITH CO-...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Chen, Hui Zhang, Hanyi Wang, Longbiao Lee, Kong Aik Liu, Meng Dang, Jianwu Tianjin International Engineering Institute Tianjin University Tianjin China Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China Institute for Infocomm Research A*STAR Singapore

In self-supervised speaker verification, the quality of pseudo labels determines the upper bound of its performance and it is not uncommon to end up with massive amount of unreliable pseudo labels. We observe that the complementary information in different modalities ensures a robust supervisory signal for audio and visual representation learning. This motivates us to propose an audio-visual self-supervised learning framework named Co-Meta Learning. Inspired by the Co-teaching+, we design a strategy that allows the information of two modalities to be coordinated through the Update by Disagreement. Moreover, we use the idea of model-agnostic meta learning (MAML) to update the network parameters, which makes the hard samples of two modalities to be better resolved by the other modality through gradient regularization. Compared to the baseline, our proposed method achieves a 29.8%, 11.7% and 12.9% relative improvement on Vox-O, Vox-E and Vox-H trials of Voxceleb1 evaluation dataset respectively. ©2023 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

TUT4CRS: Time-aware User-preference Tracking for Conversational Recommendation System 24

TUT4CRS: Time-aware User-preference Tracking for Conversatio...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： He, Dongxiao Zhang, Jinghan Wang, Xiaobao Ge, Meng Feng, Zhiyong Wang, Longbiao Ma, Xiaoke Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China Saw Swee Hock School of Public Health National University of Singapore Singapore Singapore School of Computer Science and Technology Xidian University Xi'an China

ISBN: (纸本)9798400706868

The Conversational Recommendation System (CRS) aims to capture user dynamic preferences and provide item recommendations based on multi-turn conversations. However, effectively modeling these dynamic preferences faces challenges due to conversational limitations, which mainly manifests as limited turns in a conversation (quantity aspect) and low compliance with queries (quality aspect). Previous studies often address these challenges in isolation, overlooking their interconnected nature. The fundamental issue underlying both problems lies in the potential abrupt changes in user preferences, to which CRS may not respond promptly. We acknowledge that user preferences are influenced by temporal factors, serving as a bridge between conversation quantity and quality. Therefore, we propose a more comprehensive CRS framework called Time-aware User-preference Tracking for Conversational Recommendation System (TUT4CRS), leveraging time dynamics to tackle both issues simultaneously. Specifically, we construct a global time interaction graph to incorporate rich external information and establish a local time-aware weight graph based on this information to adeptly select queries and effectively model user dynamic preferences. Extensive experiments on two real-world datasets validate that TUT4CRS can significantly improve recommendation performance while reducing the number of conversation turns. © 2024 ACM.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

A Prompt Learning Framework with Large Language Model Augmentation for Few-shot Multi-label Intent Detection

A Prompt Learning Framework with Large Language Model Augmen...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhuang, Ning Wei, Xiao Li, Junlei Wang, Xiaobao Wang, Chenyang Wang, Longbiao Dang, Jianwu Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China Shenzhen China Co. Ltd. Tianjin China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

ISBN: (纸本)9798350368741

Intent detection (ID) is essential in spoken language understanding, especially in multi-label settings where intent labels are interdependent and diverse. Existing methods like SE-MLP and QA-FT struggle in few-shot settings, due to limited data availability and efficiency concerns. To address this, we introduce a Prompt Learning framework with large language Model Augmentation (PLMA) for few-shot multi-label ID. In this study, we make three contributions. First, PLMA integrates large language models (LLMs) with small language models (SLMs), using LLMs to enhance SLMs, with prompt learning as the core framework. Second, it leverages LLMs to improve the model's understanding of both queries and labels by intent spans extraction and answer space expansion. Third, PLMA refines the QA-FT. A novel QA format template is designed to allow single-step training and inference for each input, improving efficiency. Experiments on NLU++ show that PLMA significantly outperforms other baselines in both in-domain and cross-domain settings, demonstrating its effectiveness in combining the strengths of both small and large models. © 2025 IEEE.

关键词： large language model multi-label intent detection prompt learning spoken language understanding

来源：评论

学校读者我要写书评

暂无评论

Cross-Modal Audio-Visual Co-Learning for Text-Independent Speaker Verification 48

Cross-Modal Audio-Visual Co-Learning for Text-Independent Sp...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Liu, Meng Lee, Kong Aik Wang, Longbiao Zhang, Hanyi Zeng, Chang Dang, Jianwu Tianjin University College of Intelligence and Computing Tianjin Key Laboratory of Cognitive Computing and Application Tianjin China Astar Institute for Infocomm Research Singapore Singapore Institute of Technology Singapore National Institute of Informatics Tokyo Japan

ISBN: (纸本)9781728163277

Visual speech (i.e., lip motion) is highly related to auditory speech due to the co-occurrence and synchronization in speech production. This paper investigates this correlation and proposes a cross-modal speech co-learning paradigm. The primary motivation of our cross-modal co-learning method is modeling one modality aided by exploiting knowledge from another modality. Specifically, two cross-modal boosters are introduced based on an audio-visual pseudo-siamese structure to learn the modality-transformed correlation. Inside each booster, a max-feature-map embedded Transformer variant is proposed for modality alignment and enhanced feature generation. The network is co-learned both from scratch and with pretrained models. Experimental results on the test scenarios demonstrate that our proposed method achieves around 60% and 20% average relative performance improvement over baseline unimodal and fusion systems, respectively. © 2023 IEEE.

关键词： co-learning cross-modal lip biometrics speaker verification visual speech

来源：评论

学校读者我要写书评

暂无评论

Characterising Effective Connectivity Changes in Patients with Hepatic Encephalopathy after Liver Transplantation: A Spectral Dynamic Casual Modeling Study 23

Characterising Effective Connectivity Changes in Patients wi...

引用

Proceedings of the 2023 4th International Symposium on Artificial Intelligence for Medicine Science

作者： Zhouhanyu Shen Jinghan Ouyang Yue Cheng Xiaodong Zhang Junhai Xu College of Intelligence and Computing Tianjin Key Laboratory of Cognitive Computing and Application Tianjin University China College of Intelligence and Computing Tianjin Key Laboratory of Cognitive Computing and Application Tianjin University China and Department of Radiology Tianjin First Central Hospital China College of Intelligence and Computing Tianjin Key Laboratory of Cognitive Computing and Application Tianjin University China and China Xiongan Group Digital City Technology Co. Ltd China

ISBN: (纸本)9798400708138

Objective: The exact effect of overt hepatic encephalopathy (OHE) on the neural processes involved in cognitive enhancement after liver transplantation (LT) is not currently known and requires additional study. The study aims to probe into whether the causal influences within the default mode network (DMN) are altered in OHE patients after LT using resting state functional MRI (rs-fMRI) *** and Methods: In this study, a total of 21 cirrhotic patients with OHE were included. Each patient underwent rs-fMRI scans both prior to and one month post LT. Four regions of the DMN (medial frontal cortex (MFC), posterior cingulate cortex (PCC), left parietal cortex (LPC), and right parietal cortex (RPC)) were defined for each patient using a group independent component analysis (ICA) approach. Spectral dynamic causal modeling (DCM) was employed to estimate the coupling parameters. The coupling parameters provide insights into the directional and strength of the interconnections between the DMN regions, which represent the causal interactivities among the ***: After one month of LT, the influences from the PCC to RPC, RPC to PCC and from LPC to MFC are markedly ***: The findings indicate that the disrupted causal influences including RPC exerts on PCC, PCC exerts on RPC and LPC exerts on MFC are responsive for cirrhotic patients with OHE to operate LT. And these changes in the effective connectivity have the potential to be utilized as neuroimaging biomarkers, aiding in the assessment of cognitive function restoration post LT in patients with liver cirrhosis.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Infusing Hierarchical Guidance into Prompt Tuning: A Parameter-Efficient Framework for Multi-level Implicit Discourse Relation Recognition

arXiv

引用

arXiv 2024年

作者： Zhao, Haodong He, Ruifang Xiao, Mengnan Xu, Jing College of Intelligence and Computing Tianjin University Tianjin China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin China

Multi-level implicit discourse relation recognition (MIDRR) aims at identifying hierarchical discourse relations among arguments. Previous methods achieve the promotion through fine-tuning PLMs. However, due to the data scarcity and the task gap, the pre-trained feature space cannot be accurately tuned to the task-specific space, which even aggravates the collapse of the vanilla space. Besides, the comprehension of hierarchical semantics for MIDRR makes the conversion much harder. In this paper, we propose a prompt-based Parameter-Efficient Multilevel IDRR (PEMI) framework to solve the above problems. First, we leverage parameter-efficient prompt tuning to drive the inputted arguments to match the pre-trained space and realize the approximation with few parameters. Furthermore, we propose a hierarchical label refining (HLR) method for the prompt verbalizer to deeply integrate hierarchical guidance into the prompt tuning. Finally, our model achieves comparable results on PDTB 2.0 and 3.0 using about 0.1% trainable parameters compared with baselines and the visualization demonstrates the effectiveness of our HLR method. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Co-training with Progressive Distribution Alignment and Uncertainty-Interactive Relabeling for Semi-Supervised Domain Adaptive Semantic Segmentation

Co-training with Progressive Distribution Alignment and Unce...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Ruiguo Yu Yida Wang Xuewei Li Xuzhou Fu Zijian Zhang Yuan Tian Jie Gao College of Intelligence and Computing Tianjin University China Tianjin Key Laboratory of Cognitive Computing and Application China Tianjin Key Laboratory of Advanced Networking China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Self-training is a strong baseline for semi-supervised domain adaptive semantic segmentation. However, it inevitably introduces biased links between features and concepts in the prediction of certain "hard pixels", which may mislead the generalization of models. We consider these hard pixels to come from two aspects: First, the labels are severely imbalanced and distributed across domains and classes, which may lead to features biased towards source domains and majority classes. Second, the naive threshold filtering pseudo-label methods limit the supervision of hard pixels. To address the above problems, we propose a novel co-training framework with progressive distribution alignment and uncertainty-interactive relabeling strategies. More concretely, a progressive distribution alignment strategy is proposed to match distribution across domains while providing additional supervision for tail class pixels. Additionally, an uncertainty-interactive relabeling strategy is proposed to retain more supervisory information for hard pixels and reduce the overall uncertainty of the pseudo-labels. Experiments on two widely-used benchmarks demonstrate the effectiveness of the proposed PDAUR, achieving state-of-the-art results.

关键词： Adaptation models Uncertainty Filtering Semantic segmentation Semantics Tail Signal processing Predictive models Benchmark testing Speech processing

来源：评论

学校读者我要写书评

暂无评论

Implicit Feature Augmentation with Feature Transfer For Class-Imbalanced Medical Image Classification

Implicit Feature Augmentation with Feature Transfer For Clas...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Yu, Mei Zheng, Hao Li, Xuewei Gao, Jie Fu, Xuzhou Liu, Zhiqiang Yu, Ruiguo College of Intelligence and Computing Tianjin University Tianjin300350 China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin300350 China Tianjin Key Laboratory of Advanced Networking Tianjin300350 China School of Feature Technology Tianjin University Tianjin300350 China

ISBN: (纸本)9798350337488

The class imbalance problem, which is prevalent in medical image datasets, seriously affects the diagnostic effectiveness of deep learning-based network models. To alleviate this problem, data re-sampling and loss re-weighting techniques are often used to reshape the decision boundary of the classifier. However, these techniques still lead to biased decision boundary due to the lack of sufficient and diversified samples in the tail classes of medical image datasets. In this paper, we propose an implicit feature augmentation with feature transfer (FT-IFA) method which solves the class imbalance problem by expanding the feature space of tail classes to reshape the decision boundary of the classifier. FT-IFA utilizes prototype similarity to transfer the rich transformation information from the head classes to tail classes on the basis of the balanced feature space, enriching the intra-class diversity of tail classes. Experimental results on two class-imbalanced medical image datasets show that FT-IFA outperforms the current state-of-the-art methods and effectively solves the class imbalance problem. © 2023 IEEE.

关键词： Class imbalance Feature augmentation Feature transfer Medical image classification

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：