Neurocomputing最新文献_第5页

SCR: A completion-then-reasoning framework for multi-hop question answering over incomplete knowledge graph 基于不完全知识图的多跳问答的补全-推理框架

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-16 DOI: 10.1016/j.neucom.2025.131027

Ridong Han , Jia Liu , Haijia Bi , Tao Peng , Lu Liu

{"title":"SCR: A completion-then-reasoning framework for multi-hop question answering over incomplete knowledge graph","authors":"Ridong Han , Jia Liu , Haijia Bi , Tao Peng , Lu Liu","doi":"10.1016/j.neucom.2025.131027","DOIUrl":"10.1016/j.neucom.2025.131027","url":null,"abstract":"<div><div>Reinforcement learning has become the widely adopted technique for multi-hop knowledge graph question answering task thanks to its excellent interpretability in reasoning process. However, it is severely affected by the incompleteness of knowledge graphs and the sparse rewards caused by weak supervision. In this paper, we propose a completion-then-reasoning framework, called SCR, to address these two issues. For the incompleteness of knowledge graphs, we first extract a subgraph from the given knowledge graph for a given question, and use the knowledge graph embedding model to predict and complete missing triples, followed by reinforcement learning for answer reasoning on the completed subgraph. To alleviate the sparse rewards in reinforcement learning, we introduce a semantic reward based on the semantic similarity between original question and full relational path, enabling the model to receive partial rewards for partially correct paths instead of a zero reward. Detailed experiments on PQ, PQL, MetaQA, and WebQSP datasets demonstrate that our SCR model effectively improves the performance of multi-hop knowledge graph question answering task. Particularly, under sparse KG setting, SCR model outperforms baselines by a large margin, highlighting the effectiveness of completion-then-reasoning framework in mitigating the incompleteness of knowledge graphs.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 131027"},"PeriodicalIF":5.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144686694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MarIns3D: An open-vocabulary 3D instance segmentation model with mask refinement MarIns3D：一个开放词汇的3D实例分割模型

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-16 DOI: 10.1016/j.neucom.2025.131018

Haiyang Li, Jinhe Su, Dong Zhou, Mengyun Cao

{"title":"MarIns3D: An open-vocabulary 3D instance segmentation model with mask refinement","authors":"Haiyang Li, Jinhe Su, Dong Zhou, Mengyun Cao","doi":"10.1016/j.neucom.2025.131018","DOIUrl":"10.1016/j.neucom.2025.131018","url":null,"abstract":"<div><div>Open-vocabulary 3D instance segmentation has gained significant attention due to its potential role in scene perception. Existing methods typically involve two stages: generating class-agnostic 3D instance masks using segmentation models, followed by semantic classification of these masks. However, poor classification performance often stems from low-quality masks in the first stage. This paper proposes two key components to optimize the mask generation process: a dynamic offset module and a projection consistency loss. By dynamically adjusting sampling point positions, query points can capture key scene features to generate high-quality masks. Then the projection consistency loss compares these 3D instance masks with ground truth in 2D projections to refine them, improving segmentation accuracy. Experimental results on the ScanNetV2 validation set show that MarIns3D outperforms SOLE on zero-shot segmentation, with a 1.8 % and 1.7 % improvement in AP25 and AP50, respectively, and also demonstrates enhanced open-set segmentation capabilities. These results confirm our model’s superior mask quality and segmentation performance. Furthermore, ablation studies verify that the synergy between the dynamic offset module and the projection consistency loss is crucial for these enhancements.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 131018"},"PeriodicalIF":5.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144672040","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Mol-L2: Transferring text knowledge with frozen language models for molecular representation learning Mol-L2：用冻结语言模型转移文本知识用于分子表示学习

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-16 DOI: 10.1016/j.neucom.2025.130837

Maotao Liu , Qun Liu , Xu Gong , Yunsong Luo , Guoyin Wang

{"title":"Mol-L2: Transferring text knowledge with frozen language models for molecular representation learning","authors":"Maotao Liu , Qun Liu , Xu Gong , Yunsong Luo , Guoyin Wang","doi":"10.1016/j.neucom.2025.130837","DOIUrl":"10.1016/j.neucom.2025.130837","url":null,"abstract":"<div><div>How to integrate abundant chemical text descriptions to produce expressive molecular representations is a compelling challenge. In this study, we propose a deep network architecture called Mol-L2, which aims to leverage powerful language models to transfer chemical text knowledge and enhance molecular representation learning. The main novelty of this work is the use of a two-stage training pipeline to align text and chemical spaces, where stage 1 pre-trains the language model using a specially constructed multi-objective loss, and stage 2 fine-tunes on molecular captioning. Subsequently, the output of the language model encoder is converted into a fixed-length text-enhanced embedding via a lightweight mapping network. Furthermore, a dedicated encoder containing information propagation of specific functional groups is designed to generate molecular initial representation and integrated with the text-enhanced embeddings using a weighted fusion module. Finally, the enhanced molecular representation is utilized for various downstream tasks through an additional output layer. The performance of the proposed Mol-L2 is tested on several standard benchmarks for molecular machine learning, including molecular properties prediction, drug-target interaction (DTI), and drug-drug interaction (DDI). Through comprehensive experiments, we demonstrate the merits and state-of-the-art performance of the Mol-L2 framework. Take blood–brain barrier penetration prediction, for instance, where Mol-L2 achieves the smallest prediction error, while the best comparison method is 91.8%, an improvement of 3.1%.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 130837"},"PeriodicalIF":5.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144672039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An integrated multi-attribute graph sequence clustering with fuzzy information granule and knowledge-guidance 基于模糊信息颗粒和知识引导的多属性图序列集成聚类

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-16 DOI: 10.1016/j.neucom.2025.130989

Fang Li, Jingxian Ma, Xinran Cheng

{"title":"An integrated multi-attribute graph sequence clustering with fuzzy information granule and knowledge-guidance","authors":"Fang Li, Jingxian Ma, Xinran Cheng","doi":"10.1016/j.neucom.2025.130989","DOIUrl":"10.1016/j.neucom.2025.130989","url":null,"abstract":"<div><div>Attribute graph sequence clustering is a type of vertex clustering that meets the needs of applications, like port scale clustering, which involves vertex attribute information and historical information. However, the existing methods explore such clustering tend to focus on only one of these aspects, missing opportunity for integrated analysis of data characteristics. To address this gap, our study introduces fuzzy information granule (FIG) to include as many vertices as possible while capturing multi-faceted information with precision, thereby enabling the integration of attribute information and historical information simultaneously. Based on these granular knowledges, two algorithms are raised to realize attribute graph sequence clustering, respectively for single-attribute graph sequence (Algorithm 1) and multi-attribute graph sequence (Algorithm 2). Algorithm 1 clusters single-attribute graph sequence employing FIG based fuzzy C-means algorithm, with the objects of FIGs. Leveraging the results of single-attribute clustering, Algorithm 2 extends to multi-attribute graph sequence clustering according to knowledge-guided idea. Ultimately, an accurate clustering result for multi-attribute graph sequence is ensured by considering the clustering result of each individual attribute. Worth noting that two novel proposed clustering algorithms achieve the clustering of attribute graph sequence successfully at application level, and also facilitate that of FIGs at methodological level. After comparing the proposed multi-attribute graph sequence clustering algorithm with six other algorithms on seven datasets, Algorithm 2 wins the highest clustering accuracy (CA) and adjusted rand index (ARI) values, and the smallest FIG-Index value, particularly with CA=0.88, ARI=0.62, and FIG‐Index=0.17 on a synthetic dataset. These results demonstrate that the newly proposed algorithm significantly outperforms other algorithms in clustering accuracy and robustness.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 130989"},"PeriodicalIF":5.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144685493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A comprehensive review of deep learning techniques for interaction-aware trajectory prediction in urban autonomous driving 城市自动驾驶中交互感知轨迹预测的深度学习技术综述

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-16 DOI: 10.1016/j.neucom.2025.131014

Iago Pachêco Gomes, Denis Fernando Wolf

引用次数: 0

PHiD: Preserving human identity in pose-guided character animation 博士：在姿势引导的角色动画中保存人类身份

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-16 DOI: 10.1016/j.neucom.2025.130885

Wenjie Zheng, Xingze Zou, Lianrui Mu, Jing Wang, Jiaqi Hu, Jiangnan Ye, Jiedong Zhuang, Mudassar Ali, Olumayowa Idowu, Haoji Hu

{"title":"PHiD: Preserving human identity in pose-guided character animation","authors":"Wenjie Zheng, Xingze Zou, Lianrui Mu, Jing Wang, Jiaqi Hu, Jiangnan Ye, Jiedong Zhuang, Mudassar Ali, Olumayowa Idowu, Haoji Hu","doi":"10.1016/j.neucom.2025.130885","DOIUrl":"10.1016/j.neucom.2025.130885","url":null,"abstract":"<div><div>Generating pose-guided human animation videos is a challenging task, particularly in maintaining consistent facial identity (ID) between the generated video and the reference image. Despite significant advancements in diffusion-based human animation models, existing methods, which mainly rely on basic conditioning mechanisms, often struggle with facial consistency and realism, especially when the face occupies a small region in the reference. This paper proposes a facial-area-aware approach, <strong>PHiD</strong>, designed to enhance facial ID similarity while ensuring strong structural and temporal coherence. Specifically, we propose a <strong>Pose-Driven Face Morphing</strong> module that leverages the 3D Morphable Model to synthesize proxy faces based on the reference ID and target pose, generating multi-view features to enhance temporal consistency. Additionally, we introduce a <strong>Masked Face Adapter (MFA)</strong> that embeds the proxy face and employs masked attention on facial regions to capture and refine localized facial features accurately. To enable effective training of MFA, we design a <strong>Facial ID-Preserving Loss</strong> that combines feature similarity, reconstruction, and pose consistency terms. Notably, our method demonstrates strong generalization capabilities and can be seamlessly integrated into existing pose-guided image-to-video models. Extensive experiments show that our approach outperforms baseline methods in generating human animation videos with improved facial consistency and similarity.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 130885"},"PeriodicalIF":5.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144678906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Safe reinforcement learning for discrete-time nonlinear zero-sum games with unknown state constraints and asymmetric input constraints 具有未知状态约束和非对称输入约束的离散时间非线性零和博弈的安全强化学习

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-16 DOI: 10.1016/j.neucom.2025.131026

Shihan Liu , Zhi Chen , Dongxu Gao

{"title":"Safe reinforcement learning for discrete-time nonlinear zero-sum games with unknown state constraints and asymmetric input constraints","authors":"Shihan Liu , Zhi Chen , Dongxu Gao","doi":"10.1016/j.neucom.2025.131026","DOIUrl":"10.1016/j.neucom.2025.131026","url":null,"abstract":"<div><div>In this paper, we propose a novel safe reinforcement learning (RL) algorithm for discrete-time nonlinear zero-sum games with unknown state constraints and asymmetric input constraints. To address this constrained optimal problem, we adopt a value iteration framework based on neural networks, incorporating a critic-only structure. Given the unknown safety constraints, we tackle the state constraint issue by introducing a neural network-based control barrier function (CBF) using collected data to augment the reward function. Furthermore, by leveraging the non-monotonic increasing property of the value function, we ensure the system’s safety. Additionally, we construct a non-quadratic function to further augment the reward function, thereby satisfying the asymmetric input constraints. This paper also includes a series of theoretical proofs that rigorously demonstrate the convergence and safety of the proposed algorithm. Finally, experiments conducted under different scenarios and parameter settings, compared with existing algorithms, validate the algorithm’s effectiveness and safety.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 131026"},"PeriodicalIF":5.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144686696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrating large foundation models into multimodal named entity recognition with evidential fusion 基于证据融合将大型基础模型集成到多模态命名实体识别中

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-15 DOI: 10.1016/j.neucom.2025.131015

Weide Liu , Xiaoyang Zhong , Jingwen Hou , Shaohua Li , Haozhe Huang , Wei Zhou , Yuming Fang

{"title":"Integrating large foundation models into multimodal named entity recognition with evidential fusion","authors":"Weide Liu , Xiaoyang Zhong , Jingwen Hou , Shaohua Li , Haozhe Huang , Wei Zhou , Yuming Fang","doi":"10.1016/j.neucom.2025.131015","DOIUrl":"10.1016/j.neucom.2025.131015","url":null,"abstract":"<div><div>Multimodal Named Entity Recognition (MNER) is a crucial task for information extraction from social media platforms such as Twitter. Most current methods rely on attention weights to extract information from both text and images but are often unreliable and lack interpretability. To address this problem, we propose incorporating uncertainty estimation into the MNER task, producing trustworthy predictions. Our proposed algorithm models the distribution of each modality as a Normal-inverse Gamma distribution, and fuses them into a unified distribution with an evidential fusion mechanism, enabling hierarchical characterization of uncertainties and promotion of prediction accuracy and trustworthiness. Additionally, we explore the potential of pre-trained large foundation models in MNER and propose an efficient fusion approach that leverages their robust feature representations. Experiments on two datasets demonstrate that our proposed method outperforms the baselines and achieves new state-of-the-art performance. Our code is available at <span><span>https://github.com/ZhongAobo/evi-mner</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"652 ","pages":"Article 131015"},"PeriodicalIF":5.5,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144704786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DeInfoReg: A decoupled learning framework with information regularization for better training throughput DeInfoReg：一个具有信息正则化的解耦学习框架，用于更好的训练吞吐量

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-15 DOI: 10.1016/j.neucom.2025.130813

Zih-Hao Huang, You-Teng Lin, Hung-Hsuan Chen

引用次数: 0

Trade-off analysis between finite-time synchronization and energy consumption for fractional-order two-layer neural networks 分数阶双层神经网络有限时间同步与能量消耗的权衡分析

IF 5.5 2区计算机科学

Neurocomputing Pub Date : 2025-07-15 DOI: 10.1016/j.neucom.2025.130872

Tian Lan , Baoxian Wang , Jigui Jian , Kai Wu

{"title":"Trade-off analysis between finite-time synchronization and energy consumption for fractional-order two-layer neural networks","authors":"Tian Lan , Baoxian Wang , Jigui Jian , Kai Wu","doi":"10.1016/j.neucom.2025.130872","DOIUrl":"10.1016/j.neucom.2025.130872","url":null,"abstract":"<div><div>This paper investigates the finite-time synchronization (FTS) of fractional-order two-layer neural networks with time delays and explores the energy consumption of their controllers, as well as the trade-off between the synchronization time cost and the controller energy consumption. The study makes the following key contributions: First, a sufficient criterion is derived to guarantee the FTS of fractional-order delayed complex networks by using a lemma based on fractional-order differential inequalities. This approach circumvents a potential methodological flaw present in some prior studies. Additionally, while existing research on FTS has primarily focused on fractional-order complex networks without time delays, this work extends the analysis to systems with time delays, thereby broadening the applicability of the results. Second, a switching controller is proposed to accelerate the synchronization of the error system. Third, a standardized evaluation function is introduced to analyze the trade-off between the FTS time cost and controller energy consumption in fractional-order systems, whereas previous research in this area has mostly focused on integer-order systems. Finally, the validity of the theoretical results is demonstrated through numerical simulations.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"651 ","pages":"Article 130872"},"PeriodicalIF":5.5,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0