检索结果-内蒙古大学图书馆

31st International Conference on Computational Linguistics, COLING 2025

作者： Zhao, Feng Zhang, Zhilu Yan, Cheng Liu, Xianggan Natural Language Processing and Knowledge Graph Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9798891761964

In knowledge graphs (KGs), predicting missing relations is a critical reasoning task. Recent subgraph-based models have delved into inductive settings, which aim to predict relations between newly added entities. While these models have demonstrated the ability for inductive reasoning, they only consider the structural information of the subgraph and neglect the loss of semantic information caused by replacing entities with nodes. To address this problem, we propose a novel Commonsense Subgraph Meta-Learning (CSML) model. Specifically, we extract concepts from entities, which can be viewed as high-level semantic information. Unlike previous methods, we use concepts instead of nodes to construct commonsense subgraphs. By combining these with structural subgraphs, we can leverage both structural and semantic information for more comprehensive and rational predictions. Furthermore, we regard concepts as meta-information and employ meta-learning to facilitate rapid knowledge transfer, thus addressing more complex few-shot scenarios. Experimental results confirm the superior performance of our model in both standard and few-shot inductive 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

language-Based Colorization with Sparse Attention and Multi-scale Cross-Modal Semantic Alignment 24th

Language-Based Colorization with Sparse Attention and Mult...

引用

24th International Conference on Algorithms and Architectures for Parallel processing, ICA3PP 2024

作者： Zhang, Ying Gao, Yutong Liu, Xuan Zhang, Lefeng Liu, Xianggan Jiang, Shan Ethnic Language Intelligent Analysis and Security Governance Minzu University of China Beijing China Faculty of Data Science City university of Macau China Natural Language Processing and Knowledge Graph Lab Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9789819615476

language-based colorization generates realistic and aesthetically appealing colors by leveraging the guidance of intuitive and user-friendly natural language descriptions. Previous methods in language-based image colorization face several significant challenges, including limited color richness, color bleeding, color distortions, and inconsistent color styles. These issues often arise due to the reliance on dense attention mechanisms, which can lead to an overemphasis on global features at the expense of local details. Furthermore, many feature fusion modules for text and grayscale images are not designed to achieve adequate alignment of the diverse feature representations, resulting in suboptimal colorization outcomes. In this paper, we explore a novel module that employs sparse attention to mitigate the issue of color bleeding and limited colorfulness in image colorization tasks. By introducing a sparse attention mechanism, our method enables more flexible allocation of computational resources with a focus on color awareness. Additionally, we propose a module that effectively aligns grayscale images with color descriptions, thereby improving the consistency and quality of the colorization results. This module leverages the synergy between image features and language descriptions to solve long-standing issues like color bleeding. Empirical evaluations demonstrate that our approach surpasses recent state-of-the-art techniques in both automatic and language-based colorization tasks, validating the effectiveness and robustness of our proposed method in generating high-quality, visually appealing colorized images. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Towards Information Sharing Beetle Antennae Search Optimization 24th

Towards Information Sharing Beetle Antennae Search Optimizat...

引用

24th International Conference on Algorithms and Architectures for Parallel processing, ICA3PP 2024

作者： Liu, Xuan Wang, Chenyan Liu, Wenjian Zhang, Lefeng Liu, Xianggan Gao, Yutong Key Laboratory of Ethnic Language Intelligent Analysis and Security Governance of MOE Minzu University of China Beijing China Hainan International College of Minzu University of China Li’an International Education Innovation pilot Zone Hainan China City University of Macau China Natural Language Processing and Knowledge Graph Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9789819615278

The study of bioinformatics-based evolutionary computation has long been of significant interest within the scientific community. Beetle antennae search algorithm is widely used because of its lightweight, however, it lacks information sharing due to individual iteration in the algorithm. In this paper, we propose an innovative pheromone-based beetle antennae search algorithm, which evolves from a single iterative individual in BAS algorithm to multiple parallel iterative individuals and incorporates the pheromone sharing mechanism found in ant colony optimization algorithm. Applying the pheromone-based beetle antennae search algorithm to the virtual machine placement problem in cloud computing, we find that pheromone sharing mechanism allows the PB-BAS algorithm to exhibit superior optimization capabilities and effectively avoids convergence to local optimal. To verify the performance of the algorithm, we select other alternative algorithms and conduct a large number of comparative experiments under different experimental setups, the experimental results show the effectiveness and efficiency of our algorithm. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Cross-Modal Mask and Detail Alignment for Text-Based Person Retrieval 24th

Cross-Modal Mask and Detail Alignment for Text-Based Perso...

引用

24th International Conference on Algorithms and Architectures for Parallel processing, ICA3PP 2024

作者： Guo, Ao Liu, Xuan Liu, Xianggan Hu, Bingmeng Yuan, Jie Chu, Chiawei Key Laboratory of Ethnic Language Intelligent Analysis and Security Governance Ministry of Education Minzu University of China Beijing100081 China Hainan International College of Minzu University of China Li’an International Education Innovation pilot Zone Hainan Lingshui Li572499 China Natural Language Processing and Knowledge Graph Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China City University of Macau China

ISBN: (纸本)9789819615278

Text-based person retrieval primarily aims to retrieve the images of target persons represented by a given text query. In this task, how to effectively align images and text globally and locally is an important challenge. At the same time, since multiple modalities are involved, reducing the differences between different modalities is also an important challenge. Existing work has focused more on reconstructing an image rather than the semantic consistency between the image and text modalities. Therefore, we introduce the Cross-Modal Mask and Detail Alignment (CMDA) framework to address these challenges. Under this framework, our proposed Cross-Modal Mask Alignment module (CMA) semantically aligns features generated by supplementing randomly masked image/text with another modality text/image. Additionally, to narrow the gap between the image and text modalities, we designed the Cross-Modal Detail Alignment module (CDA), which establishes connections between images and texts and facilitates interactions between these two different modalities. Experimental results show that our model exhibits outstanding performance across multiple public datasets, i.e., CUHK-PEDES, ICFG-PEDES, and RSTPReID. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Correcting language Model Bias for Text Classification in True Zero-Shot Learning 30

Correcting Language Model Bias for Text Classification in Tr...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on language Resources and Evaluation, LREC-COLING 2024

作者： Zhao, Feng Wan, Xianlin Yan, Cheng Loo, Chu Kiong Natural Language Processing and Knowledge Graph Lab School of Computer Science and Technology Huazhong University of Science and Technology China Faculty of Computer Science & Information Technology University of Malaya Malaysia

ISBN: (纸本)9782493814104

Combining pre-trained language models (PLMs) and manual templates is a common practice for text classification in zero-shot scenarios. However, the effect of this approach is highly volatile, ranging from random guesses to near state-of-the-art results, depending on the quality of the manual templates. In this paper, we show that this instability stems from the fact that language models tend toward predicting certain label words of text classification, and manual templates can influence this tendency. To address this, we develop a novel pipeline for annotating and filtering a few examples from unlabeled examples. Moreover, we propose a new method to measure model bias on label words that utilizes unlabeled examples as a validation set when tuning language models. Our approach does not require any pre-labeled examples. Experimental results on six text classification tasks demonstrate that the proposed approach significantly outperforms standard prompt learning in zero-shot settings, achieving up to 19.7% absolute improvement and 13.8% average improvement. More surprisingly, on IMDB and SST-2, our approach even exceeds all few-shot baselines. © 2024 ELRA language Resource Association: CC BY-NC 4.0.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

引用

2nd Joint Conference on Lexical and Computational Semantics, *SEM 2013

作者： Zesch, Torsten Levy, Omer Gurevych, Iryna Dagan, Ido Ubiquitous Knowledge Processing Lab Computer Science Department Technische Universität Darmstadt Germany Natural Language Processing Lab Computer Science Department Bar-Ilan University Israel

ISBN: (纸本)9781937284497

Our system combines text similarity measures with a textual entailment system. In the main task, we focused on the influence of lexicalized versus unlexicalized features, and how they affect performance on unseen questions and domains. We also participated in the pilot partial entailment task, where our system significantly outperforms a strong baseline. c 2013 Association for Computational Linguistics

关键词：

来源：评论

学校读者我要写书评

暂无评论

Recognizing partial textual entailment

Recognizing partial textual entailment

引用

51st Annual Meeting of the Association for Computational Linguistics, ACL 2013

作者： Levy, Omer Zesch, Torsten Dagan, Ido Gurevych, Iryna Natural Language Processing Lab. Computer Science Department Bar-Ilan University Israel Ubiquitous Knowledge Processing Lab. Computer Science Department Technische Universität Darmstadt Germany

ISBN: (纸本)9781937284510

Textual entailment is an asymmetric relation between two text fragments that describes whether one fragment can be inferred from the other. It thus cannot capture the notion that the target fragment is "almost entailed" by the given text. The recently suggested idea of partial textual entailment may remedy this problem. We investigate partial entailment under the faceted entailment model and the possibility of adapting existing textual entailment methods to this setting. Indeed, our results show that these methods are useful for recognizing partial entailment. We also provide a preliminary assessment of how partial entailment may be used for recognizing (complete) textual entailment. © 2013 Association for Computational Linguistics.

关键词： Text processing

来源：评论

学校读者我要写书评

暂无评论

Causal graph Representation Learning for Outcome-Oriented Link Prediction

Causal Graph Representation Learning for Outcome-Oriented Li...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Langjunqing Jin Feng Zhao Cheng Yan Xiangyu Gui School of Computer Science and Technology Natural Language Processing and Knowledge Graph Lab Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Causal graph representation learning has been an essential tool for improving the performance achieved in downstream tasks such as link prediction in causal graphs. Most of the existing methods can learn the links between binary nodes in an observed graph structure. However, the causal relations between nodes and their multivariate continuous nature are easily overlooked when learning knowledge graph links. This can lead to difficulty when attempting to find the real causes affecting the facts in multiple node chains, thus reducing the accuracy of prediction. Therefore, we propose a causal representation learning model for link prediction (CRLLP) to accurately complete missing links in causal graphs. Treating the node embeddings in causal graphs as contexts is a novel paradigm for characterizing causal relations in which the exsiting graph structure are treatments and the links to be completed are outcomes. In addition, we propose an outcome-oriented counterfactual link prediction method that learns causal chain representations based on observed nodes and queries the causal links in causal graphs by answering counterfactual questions. The developed method can clearly identify the causes corresponding to different outcomes. The experimental results show that our model can achieve optimal link prediction and causal chain inference performance with highly accurate causal characterization representations. Furthermore, other experiments demonstrate the superiority and interpretability of our model.

关键词： Representation learning Accuracy Neural networks knowledge graphs Predictive models Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Enabling natural Zero-Shot Prompting on Encoder Models via Statement-Tuning

arXiv

引用

arXiv 2024年

作者： Elshabrawy, Ahmed Huang, Yongxin Gurevych, Iryna Aji, Alham Fikri Department Natural Language Processing MBZUAI United Arab Emirates Ubiquitous Knowledge Processing Lab Technical University of Darmstadt Germany

While Large language Models (LLMs) exhibit remarkable capabilities in zero-shot and few-shot scenarios, they often require computationally prohibitive sizes. Conversely, smaller Masked language Models (MLMs) like BERT and RoBERTa achieve state-of-the-art results through fine-tuning but struggle with extending to few-shot and zero-shot settings due to their architectural constraints. Hence, we propose Statement-Tuning, a technique that models discriminative tasks as a set of finite statements and trains an encoder model to discriminate between the potential statements to determine the label. We do Statement-Tuning on multiple tasks to enable cross-task generalization. Experimental results demonstrate that Statement-Tuning achieves competitive performance compared to state-of-the-art LLMs with significantly fewer parameters. Moreover, the study investigates the impact of several design choices on few-shot and zero-shot generalization, revealing that Statement-Tuning can achieve strong performance with modest training data and benefits from task and statement diversity for unseen task generalizability. We release all the code used to generate statement data, train and evaluate our Statement-Tuned models. © 2024, CC BY.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

IE-Evo: Internal and External Evolution-Enhanced Temporal knowledge graph Forecasting

IE-Evo: Internal and External Evolution-Enhanced Temporal Kn...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Kangzheng Liu Feng Zhao Guandong Xu Shiqing Wu Natural Language Processing and Knowledge Graph Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Data Science and Machine Intelligence Lab University of Technology Sydney Sydney Australia

Temporal knowledge graph (TKG) forecasting is widely used in various fields due to its ability to infer future events based on historical information. Modeling the internal structures and chronological dependencies of historical subgraph sequences has been proven effective. Nevertheless, on the one hand, the TKG forecasting process generally suffers from a lack of sufficient sample data due to historical resource limitations; thus, most works focus on continuously mining the patterns of historical sequences while ignoring the semantically-rich background information provided by external knowledge, especially when historical query-related information is scarce. On the other hand, when merely serializing the given subgraph sequence to mimic its temporal evolution process, only the chronological dependencies between the subgraphs can be considered, thus ignoring the evolution of time information. Hence, a method that integrates internal and external knowledge to enhance the representations of entities is urgently needed. To this end, we propose a novel TKG forecasting method, namely, the internal and external evolution-enhanced framework (IE-Evo). For the former issue, we design an external evolution encoder and use a pre-trained language model (PLM) to provide powerful external knowledge semantics for TKG forecasting. To address the latter concern, we propose an internal evolution encoder that explicitly embeds the time information while modeling the aggregation and evolution processes of the observed sequential structural information. IE-Evo has been evaluated on four public benchmark datasets, showcasing its significant improvements across multiple evaluation metrics.

关键词：

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：