Neurocomputing最新文献_第4页

The double-edged sword: A critical review of foundational medical datasets for AI benchmarks, biases, and the future of equitable healthcare 双刃剑：对人工智能基准、偏见和公平医疗的未来的基础医疗数据集进行批判性审查

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-05 DOI: 10.1016/j.neucom.2026.132919

Rabie A. Ramadan , Nadim K.M. Madi , Sallam O.F. Khairy , Kamal Aldin Yousif , Muataz Salam Al-Daweri , Alrajhi Waleed Khalid

The advancement of Artificial Intelligence (AI) has revolutionized medical diagnostics and treatment. Large-scale public datasets are fueling research in this field. Therefore, this systematic review is a comprehensive analysis of 13 foundational medical datasets. It evaluates the characteristics, performance metrics, and inherent biases of datasets across medical imaging, electronic health records, and genomics. The published literature is systematically reviewed to categorize these datasets, with a focus on performance metrics for everyday machine learning tasks. Additionally, this research documents evidence of systemic bias and limitations that affect model generalizability and clinical equity. Our analysis reveals compelling evidence that significant limitations temper the remarkable progress of algorithms. It has been frequently observed that AI models suffer dramatic accuracy drops when tested beyond their training distribution, with the Area Under the Curve consistently declining from 0.95 to 0.63. The research also identified consistent patterns of systemic bias that threaten the equitable application of healthcare. This bias stems from unrepresentative sampling, subjective annotation practices, label noise, and Natural Language Processing-derived ground-truth labels. Our findings demonstrate the urgent need for a paradigm shift in the development of medical applications. The AI and medical communities must prioritize generating diverse datasets and mitigating systematic bias. This study provides evidence-based recommendations and a technical toolkit to address these challenges and reduce any health disparities.

人工智能（AI）的进步彻底改变了医疗诊断和治疗。大规模的公共数据集正在推动这一领域的研究。因此，本系统综述是对13个基础医学数据集的综合分析。它评估了医学成像、电子健康记录和基因组学数据集的特征、性能指标和固有偏差。系统地回顾已发表的文献，对这些数据集进行分类，重点关注日常机器学习任务的性能指标。此外，本研究记录了影响模型普遍性和临床公平性的系统性偏见和局限性的证据。我们的分析揭示了令人信服的证据，表明显著的局限性抑制了算法的显著进步。人们经常观察到，当测试超出其训练分布时，人工智能模型的准确性会急剧下降，曲线下面积（Area Under the Curve）从0.95持续下降到0.63。该研究还确定了系统性偏见的一致模式，威胁到医疗保健的公平应用。这种偏差源于非代表性采样、主观注释实践、标签噪声和自然语言处理派生的基础真值标签。我们的研究结果表明，迫切需要在医学应用的发展模式转变。人工智能和医学界必须优先考虑生成多样化的数据集和减轻系统性偏见。这项研究提供了基于证据的建议和技术工具包，以应对这些挑战并减少任何健康差距。

{"title":"The double-edged sword: A critical review of foundational medical datasets for AI benchmarks, biases, and the future of equitable healthcare","authors":"Rabie A. Ramadan , Nadim K.M. Madi , Sallam O.F. Khairy , Kamal Aldin Yousif , Muataz Salam Al-Daweri , Alrajhi Waleed Khalid","doi":"10.1016/j.neucom.2026.132919","DOIUrl":"10.1016/j.neucom.2026.132919","url":null,"abstract":"<div><div>The advancement of Artificial Intelligence (AI) has revolutionized medical diagnostics and treatment. Large-scale public datasets are fueling research in this field. Therefore, this systematic review is a comprehensive analysis of 13 foundational medical datasets. It evaluates the characteristics, performance metrics, and inherent biases of datasets across medical imaging, electronic health records, and genomics. The published literature is systematically reviewed to categorize these datasets, with a focus on performance metrics for everyday machine learning tasks. Additionally, this research documents evidence of systemic bias and limitations that affect model generalizability and clinical equity. Our analysis reveals compelling evidence that significant limitations temper the remarkable progress of algorithms. It has been frequently observed that AI models suffer dramatic accuracy drops when tested beyond their training distribution, with the Area Under the Curve consistently declining from 0.95 to 0.63. The research also identified consistent patterns of systemic bias that threaten the equitable application of healthcare. This bias stems from unrepresentative sampling, subjective annotation practices, label noise, and Natural Language Processing-derived ground-truth labels. Our findings demonstrate the urgent need for a paradigm shift in the development of medical applications. The AI and medical communities must prioritize generating diverse datasets and mitigating systematic bias. This study provides evidence-based recommendations and a technical toolkit to address these challenges and reduce any health disparities.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"675 ","pages":"Article 132919"},"PeriodicalIF":6.5,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146147239","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

Memory recall-driven multi-view semantic inference for offensive language detection 基于记忆回忆的多视图语义推理攻击性语言检测

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-05 DOI: 10.1016/j.neucom.2026.132948

Zhiqiang Zhang , Tianpeng Cheng , Bing Li , Yuankang Sun , Chengxu Wang

The detection of offensive language plays a critical role in maintaining the health of online communities, preventing cyberbullying, and fostering inclusive communication. Current approaches utilize facilitated LLMs for direct aggressiveness classification, but flaws in complex contextual reasoning and in the detection of subtle cues in conversational environments greatly reduce detection performance. To address the aforementioned challenges, we propose the Memory Recall-Driven Multi-View Semantic Inference (MR-MVSI) model. Specifically, we first build a multi-view semantic inference module that enables the model to effectively capture subtle contextual cues and underlying emotional features from situational backgrounds, communicative targets, and emotions. Meanwhile, we employ a self-check mechanism to discriminate and regenerate the generated information, thereby ensuring the rigor and reliability of the inference process. In addition, we introduce a training memory recall module, which embeds the input samples into a highly semantic space and retrieves the most relevant memory segments to interpret complex linguistic patterns, thus significantly improving the detection accuracy. The experimental results demonstrate that our proposed MR-MVSI model achieves superior performance across all three benchmark datasets (OLID, HateXplain, and HatEval), with performance improvements of

6.6 %

,

0.2 %

, and

7.6 %

respectively.

攻击性语言的检测对于维护网络社区的健康、防止网络欺凌、促进包容性交流具有至关重要的作用。目前的方法利用便利的llm进行直接攻击性分类，但复杂的上下文推理和会话环境中微妙线索的检测缺陷大大降低了检测性能。为了解决上述问题，我们提出了记忆回忆驱动的多视图语义推理（MR-MVSI）模型。具体来说，我们首先构建了一个多视图语义推理模块，使模型能够有效地从情景背景、交际目标和情绪中捕获微妙的上下文线索和潜在的情感特征。同时，我们采用自检机制对生成的信息进行判别和再生，从而保证了推理过程的严谨性和可靠性。此外，我们引入了训练记忆召回模块，该模块将输入样本嵌入到高度语义空间中，并检索最相关的记忆片段来解释复杂的语言模式，从而显着提高了检测精度。实验结果表明，我们提出的MR-MVSI模型在所有三个基准数据集（OLID， HateXplain和HatEval）上都取得了优异的性能，性能分别提高了6.6%，0.2%和7.6%。

{"title":"Memory recall-driven multi-view semantic inference for offensive language detection","authors":"Zhiqiang Zhang , Tianpeng Cheng , Bing Li , Yuankang Sun , Chengxu Wang","doi":"10.1016/j.neucom.2026.132948","DOIUrl":"10.1016/j.neucom.2026.132948","url":null,"abstract":"<div><div>The detection of offensive language plays a critical role in maintaining the health of online communities, preventing cyberbullying, and fostering inclusive communication. Current approaches utilize facilitated LLMs for direct aggressiveness classification, but flaws in complex contextual reasoning and in the detection of subtle cues in conversational environments greatly reduce detection performance. To address the aforementioned challenges, we propose the <strong>M</strong>emory <strong>R</strong>ecall-Driven <strong>M</strong>ulti-<strong>V</strong>iew <strong>S</strong>emantic <strong>I</strong>nference (MR-MVSI) model. Specifically, we first build a multi-view semantic inference module that enables the model to effectively capture subtle contextual cues and underlying emotional features from situational backgrounds, communicative targets, and emotions. Meanwhile, we employ a self-check mechanism to discriminate and regenerate the generated information, thereby ensuring the rigor and reliability of the inference process. In addition, we introduce a training memory recall module, which embeds the input samples into a highly semantic space and retrieves the most relevant memory segments to interpret complex linguistic patterns, thus significantly improving the detection accuracy. The experimental results demonstrate that our proposed MR-MVSI model achieves superior performance across all three benchmark datasets (OLID, HateXplain, and HatEval), with performance improvements of <span><math><mn>6.6</mn><mi>%</mi></math></span>, <span><math><mn>0.2</mn><mi>%</mi></math></span>, and <span><math><mn>7.6</mn><mi>%</mi></math></span> respectively.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"675 ","pages":"Article 132948"},"PeriodicalIF":6.5,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146147512","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

MODE+: A benchmark and a probe into multimodal open-domain dialogue evaluation MODE+：对多模态开放域对话评估的基准和探索

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-05 DOI: 10.1016/j.neucom.2026.132787

Hang Yin , Xinglin Wang , Yueqi Zhang, Pinren Lu, Bin Sun, Peiwen Yuan, Kan Li

Multimodal Open-domain Dialogue (MOD) plays a crucial role in AI-human interactions and has garnered substantial interest. Although existing studies have explored various aspects of MOD, the evaluation of MOD remains underexplored. In this work, we propose MODE+, an evaluation benchmark for MOD and a probe into multimodel open-domain dialogue evaluation. Specifically, we construct MODE with a balanced difficulty distribution and divide it into three parts: MODE-Base and MODE-Hard, both consisting of single-turn dialogues, with MODE-Base containing 889 test cases and MODE-Hard comprising 215 more challenging cases designed for probing model robustness against multimodal inconsistencies. Additionally, we include MODE-Multi, which contains over 10,000 multi-turn dialogue cases for more extensive testing. Each case contains an image, a context, and turn-level response scores provided by at least three human annotators following standardized criteria. The consistency of human annotations has an average Spearman correlation of over 0.9, indicating that MODE is highly reliable in annotation. We test the MOD evaluation capabilities of various evaluators on MODE, including LLaMA, Claude3, GPT-4, LLaVA, Gemini and Qwen3-VL. Results show that even the best-performing model-based evaluators have surprisingly low agreement with human evaluations, with consistency scores for MODE-Base below 0.7 and for MODE-Hard falling below 0.4. To improve model-based MOD evaluation capabilities, we propose the MM-Eval framework, a systematic methodology designed to standardize automatic evaluation. MM-Eval introduces Image Transformation as a modality-bridging mechanism, Inference Enhancement for transparent reasoning, and Inference Calibration for statistical reliability. Compared to the baselines, MM-Eval achieves a 67.41% improvement on MODE-Base and a 297% enhancement on MODE-Hard. Furthermore, the performance on MODE-Multi shows significant improvements with MM-Eval, demonstrating that the framework is capable of handling larger and more complex datasets. These results demonstrate that MM-Eval serves as a transferable and robust standard for future MOD evaluation.

多模态开放域对话（multi - modal Open-domain Dialogue， MOD）在人工智能与人类交互中起着至关重要的作用，已经引起了人们的广泛关注。虽然已有的研究已经对MOD的各个方面进行了探索，但对MOD的评价仍存在不足。在这项工作中，我们提出了MOD的评估基准MODE+，并对多模型开放域对话评估进行了探索。具体来说，我们构建了具有平衡难度分布的MODE，并将其分为三部分：MODE- base和MODE- hard，均由单回合对话组成，MODE- base包含889个测试用例，MODE- hard包含215个更具挑战性的用例，旨在探索模型对多模态不一致性的鲁棒性。此外，我们还包括MODE-Multi，它包含超过10,000个多回合对话案例，用于更广泛的测试。每个案例包含一个图像、一个上下文和由至少三个人类注释者按照标准化标准提供的回合级响应分数。人工标注的一致性平均Spearman相关系数大于0.9，说明MODE在标注上具有较高的可靠性。我们测试了包括LLaMA、Claude3、GPT-4、LLaVA、Gemini和Qwen3-VL在内的多种评估器在MODE上的MOD评估能力。结果表明，即使是表现最好的基于模型的评估器与人类评估的一致性也低得惊人，MODE-Base的一致性得分低于0.7，MODE-Hard的一致性得分低于0.4。为了提高基于模型的MOD评估能力，我们提出了MM-Eval框架，这是一种系统化的方法，旨在规范自动评估。MM-Eval引入了图像转换作为模态桥接机制，用于透明推理的推理增强和用于统计可靠性的推理校准。与基线相比，MM-Eval在MODE-Base上提高了67.41%，在MODE-Hard上提高了297%。此外，MM-Eval在MODE-Multi上的性能得到了显著改善，表明该框架能够处理更大、更复杂的数据集。这些结果表明，MM-Eval可作为未来国防部评估的可转移和稳健的标准。

{"title":"MODE+: A benchmark and a probe into multimodal open-domain dialogue evaluation","authors":"Hang Yin , Xinglin Wang , Yueqi Zhang, Pinren Lu, Bin Sun, Peiwen Yuan, Kan Li","doi":"10.1016/j.neucom.2026.132787","DOIUrl":"10.1016/j.neucom.2026.132787","url":null,"abstract":"<div><div>Multimodal Open-domain Dialogue (MOD) plays a crucial role in AI-human interactions and has garnered substantial interest. Although existing studies have explored various aspects of MOD, the evaluation of MOD remains underexplored. In this work, we propose MODE+, an evaluation benchmark for MOD and a probe into multimodel open-domain dialogue evaluation. Specifically, we construct MODE with a balanced difficulty distribution and divide it into three parts: MODE-Base and MODE-Hard, both consisting of single-turn dialogues, with MODE-Base containing 889 test cases and MODE-Hard comprising 215 more challenging cases designed for probing model robustness against multimodal inconsistencies. Additionally, we include MODE-Multi, which contains over 10,000 multi-turn dialogue cases for more extensive testing. Each case contains an image, a context, and turn-level response scores provided by at least three human annotators following standardized criteria. The consistency of human annotations has an average Spearman correlation of over 0.9, indicating that MODE is highly reliable in annotation. We test the MOD evaluation capabilities of various evaluators on MODE, including LLaMA, Claude3, GPT-4, LLaVA, Gemini and Qwen3-VL. Results show that even the best-performing model-based evaluators have surprisingly low agreement with human evaluations, with consistency scores for MODE-Base below 0.7 and for MODE-Hard falling below 0.4. To improve model-based MOD evaluation capabilities, we propose the MM-Eval framework, a systematic methodology designed to standardize automatic evaluation. MM-Eval introduces Image Transformation as a modality-bridging mechanism, Inference Enhancement for transparent reasoning, and Inference Calibration for statistical reliability. Compared to the baselines, MM-Eval achieves a 67.41% improvement on MODE-Base and a 297% enhancement on MODE-Hard. Furthermore, the performance on MODE-Multi shows significant improvements with MM-Eval, demonstrating that the framework is capable of handling larger and more complex datasets. These results demonstrate that MM-Eval serves as a transferable and robust standard for future MOD evaluation.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"675 ","pages":"Article 132787"},"PeriodicalIF":6.5,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146147508","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

Density-increment and cut-edge optimized clustering via minimum spanning forest 基于最小生成森林的密度增量与边缘优化聚类

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-05 DOI: 10.1016/j.neucom.2026.132957

Haoyu Zhai , Jie Yang , Hantao Guo , Bin Wang , Yan Ma

Graph-based clustering methods have gained significant attention due to their ability to model complex data structures and uncover intrinsic relationships. However, traditional approaches, such as the k-nearest neighbor graph and minimum spanning tree (MST), often suffer from poor connectivity and fail to accurately preserve local structures, thereby limiting clustering performance. To address these challenges, this paper proposes a novel minimum spanning forest (MSF)-based clustering algorithm that follows a multi-stage split-merge strategy. First, an MSF is constructed by iteratively generating and merging α MSTs, enabling a more robust representation of both local and global structures. In the splitting phase, a density-driven traversal is performed along the path of maximum density increment, ensuring that clusters form around density increment peaks while maintaining structural consistency. The merging phase consists of two stages: (1) an initial merging step that groups spatially adjacent small clusters and (2) a refined merging process based on a new inter-cluster distance metric that incorporates cut points and cut edges, facilitating an adaptive and topology-aware merging strategy. Extensive experiments on synthetic and UCI real datasets demonstrate that the proposed approach consistently outperforms state-of-the-art graph-based clustering methods, exhibiting superior robustness across datasets with varying densities, shapes, and noise levels.

基于图的聚类方法由于其建模复杂数据结构和揭示内在关系的能力而获得了极大的关注。然而，传统的方法，如k近邻图和最小生成树（MST），往往存在连通性差和不能准确保留局部结构的问题，从而限制了聚类性能。为了解决这些问题，本文提出了一种新的基于最小生成森林（MSF）的聚类算法，该算法遵循多阶段分裂合并策略。首先，通过迭代生成和合并α mst来构建MSF，从而使局部和全局结构具有更强的鲁棒性。在分裂阶段，沿着最大密度增量路径进行密度驱动遍历，确保簇在密度增量峰值周围形成，同时保持结构一致性。合并阶段包括两个阶段：(1)初始合并步骤，对空间上相邻的小集群进行分组；(2)基于新的集群间距离度量的细化合并过程，该度量包含切点和切边，促进自适应和拓扑感知的合并策略。在合成和UCI真实数据集上进行的大量实验表明，所提出的方法始终优于最先进的基于图的聚类方法，在不同密度、形状和噪声水平的数据集上表现出卓越的鲁棒性。

{"title":"Density-increment and cut-edge optimized clustering via minimum spanning forest","authors":"Haoyu Zhai , Jie Yang , Hantao Guo , Bin Wang , Yan Ma","doi":"10.1016/j.neucom.2026.132957","DOIUrl":"10.1016/j.neucom.2026.132957","url":null,"abstract":"<div><div>Graph-based clustering methods have gained significant attention due to their ability to model complex data structures and uncover intrinsic relationships. However, traditional approaches, such as the <em>k</em>-nearest neighbor graph and minimum spanning tree (MST), often suffer from poor connectivity and fail to accurately preserve local structures, thereby limiting clustering performance. To address these challenges, this paper proposes a novel minimum spanning forest (MSF)-based clustering algorithm that follows a multi-stage split-merge strategy. First, an MSF is constructed by iteratively generating and merging <em>α</em> MSTs, enabling a more robust representation of both local and global structures. In the splitting phase, a density-driven traversal is performed along the path of maximum density increment, ensuring that clusters form around density increment peaks while maintaining structural consistency. The merging phase consists of two stages: (1) an initial merging step that groups spatially adjacent small clusters and (2) a refined merging process based on a new inter-cluster distance metric that incorporates cut points and cut edges, facilitating an adaptive and topology-aware merging strategy. Extensive experiments on synthetic and UCI real datasets demonstrate that the proposed approach consistently outperforms state-of-the-art graph-based clustering methods, exhibiting superior robustness across datasets with varying densities, shapes, and noise levels.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"674 ","pages":"Article 132957"},"PeriodicalIF":6.5,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146192218","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

Hi-CBM: Mitigating information leakage via hierarchical concept bottleneck modeling Hi-CBM：通过分层概念瓶颈建模减少信息泄漏

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-05 DOI: 10.1016/j.neucom.2026.132916

Ao Sun , Qingyue Wang , Yuanyuan Yuan , Pingchuan Ma , Shuai Wang

Concept Bottleneck Models (CBMs) enhance interpretability and facilitate effective intervention by explicitly mapping input features to labels through human-understandable concepts. However, existing CBM frameworks often suffer from information leakage, wherein latent unintended information bypasses the concept layer, undermining interpretability and contaminating downstream predictions. To address this challenge, we propose Hi-CBM, a refined CBM framework that explicitly safeguards the two inherent mappings in CBMs—

features \to concepts

and

concepts \to classes

—to prevent leakage. Specifically, a Concept-Bottleneck Pooling mechanism regulates the feature-to-concept mapping by selectively aggregating latent features into semantic concepts to filter out irrelevant signals, while a binary Intervention Matrix governs the concept-to-class mapping by constraining concept–class associations, preventing unintended information encoded in concept probabilities from influencing final predictions. Extensive experiments across multiple datasets show that Hi-CBM substantially mitigates information leakage and produces concept representations that are both interpretable and intervenable, while maintaining strong predictive performance.

概念瓶颈模型（CBMs）通过人类可理解的概念显式地将输入特征映射到标签，从而增强了可解释性，并促进了有效的干预。然而，现有的CBM框架经常遭受信息泄漏，其中潜在的意外信息绕过概念层，破坏可解释性并污染下游预测。为了应对这一挑战，我们提出了Hi-CBM，这是一个改进的CBM框架，它明确地保护了CBM中的两个固有映射——特征→概念和概念→类——以防止泄漏。具体来说，概念-瓶颈池机制通过选择性地将潜在特征聚合到语义概念中以过滤掉无关信号来调节特征到概念的映射，而二元干预矩阵通过约束概念-类关联来控制概念到类的映射，防止概念概率中编码的意外信息影响最终预测。跨多个数据集的广泛实验表明，Hi-CBM大大减轻了信息泄漏，并产生了可解释和可干预的概念表示，同时保持了强大的预测性能。

{"title":"Hi-CBM: Mitigating information leakage via hierarchical concept bottleneck modeling","authors":"Ao Sun , Qingyue Wang , Yuanyuan Yuan , Pingchuan Ma , Shuai Wang","doi":"10.1016/j.neucom.2026.132916","DOIUrl":"10.1016/j.neucom.2026.132916","url":null,"abstract":"<div><div>Concept Bottleneck Models (CBMs) enhance interpretability and facilitate effective intervention by explicitly mapping input features to labels through human-understandable concepts. However, existing CBM frameworks often suffer from <em>information leakage</em>, wherein latent unintended information bypasses the concept layer, undermining interpretability and contaminating downstream predictions. To address this challenge, we propose Hi-CBM, a refined CBM framework that explicitly safeguards the two inherent mappings in CBMs—<span><math><mtext>features</mtext><mo>→</mo><mtext>concepts</mtext></math></span> and <span><math><mtext>concepts</mtext><mo>→</mo><mtext>classes</mtext></math></span>—to prevent leakage. Specifically, a <em>Concept-Bottleneck Pooling</em> mechanism regulates the feature-to-concept mapping by selectively aggregating latent features into semantic concepts to filter out irrelevant signals, while a binary <em>Intervention Matrix</em> governs the concept-to-class mapping by constraining concept–class associations, preventing unintended information encoded in concept probabilities from influencing final predictions. Extensive experiments across multiple datasets show that Hi-CBM substantially mitigates information leakage and produces concept representations that are both interpretable and intervenable, while maintaining strong predictive performance.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"675 ","pages":"Article 132916"},"PeriodicalIF":6.5,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146147234","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 fuzzy multi-objective neuro-evolutionary framework with bargaining-based selection for interpretable body fat prediction 基于讨价还价选择的可解释体脂预测模糊多目标神经进化框架

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-05 DOI: 10.1016/j.neucom.2026.132974

Farshid Keivanian , Niusha Shafiabady , Nasimul Noman , Zongwen Fan , Seyedali Mirjalili

Accurate prediction of body fat percentage is important for early detection of cardiovascular and metabolic risk, yet common proxies such as body mass index and bioelectrical impedance analysis often fail to capture nonlinear, high-dimensional biomedical relationships. We propose a fuzzy neuro-evolutionary multi-objective framework that optimizes feature subsets for Multi-Layer Perceptron (MLP) regression and selects clinically balanced trade-offs using a Nash-bargaining-based decision layer. Experiments on the Johnson anthropometric dataset (n = 252) and the NHANES 1999–2000 clinical dataset (n = 862) show that the single- and multi-objective variants reduce feature dimensionality by up to 65 % relative to the full feature set while achieving a 3–8 % reduction in RMSE compared with state-of-the-art fuzzy and evolutionary baselines. The proposed approach also improves stability, reflected by lower within-run residual dispersion (reported as the run-averaged value

{STD}_{err, mean}

and lower run-to-run variability (

{STD}_{runs} (.)

) over 30 independent runs; Wilcoxon signed-rank tests confirm statistically significant improvements (p < 0.05). The multi-objective model jointly optimizes parsimony (

n_{f}

), prediction accuracy (RMSE), and robustness measured by within-run residual dispersion (

{STD}_{err}

), and selects fair Pareto solutions via a cooperative Multi-Objective Bargaining Game. A symmetric V-shaped transfer function empirically improves binary neuro-evolution stability. The resulting models provide interpretable and clinically plausible predictors for body-fat estimation, supporting trustworthy biomedical decision support.

准确预测体脂百分比对于心血管和代谢风险的早期检测非常重要，但常用的替代方法，如体重指数和生物电阻抗分析，往往无法捕捉非线性、高维的生物医学关系。我们提出了一个模糊神经进化多目标框架，该框架优化了多层感知器（MLP）回归的特征子集，并使用基于纳什交易的决策层选择临床平衡权衡。在Johnson人体测量数据集（n = 252）和NHANES 1999-2000临床数据集（n = 862）上的实验表明，相对于完整的特征集，单目标和多目标变体将特征维数减少了65 %，而与最先进的模糊和进化基线相比，RMSE减少了3-8 %。所提出的方法还提高了稳定性，反映在30个独立运行中较低的运行内剩余离散度(报告为运行平均值STDerr，平均值和较低的运行间可变性（STDruns()）；Wilcoxon sign -rank检验证实了统计学上显著的改善（p < 0.05）。该多目标模型对简约性（nf）、预测精度（RMSE）和运行内剩余离散度（STDerr）的鲁棒性进行优化，并通过合作多目标议价博弈选择公平的帕累托解。对称的v形传递函数经验地提高了二元神经进化的稳定性。由此产生的模型为体脂估计提供了可解释和临床可信的预测因子，支持可靠的生物医学决策支持。

{"title":"A fuzzy multi-objective neuro-evolutionary framework with bargaining-based selection for interpretable body fat prediction","authors":"Farshid Keivanian , Niusha Shafiabady , Nasimul Noman , Zongwen Fan , Seyedali Mirjalili","doi":"10.1016/j.neucom.2026.132974","DOIUrl":"10.1016/j.neucom.2026.132974","url":null,"abstract":"<div><div>Accurate prediction of body fat percentage is important for early detection of cardiovascular and metabolic risk, yet common proxies such as body mass index and bioelectrical impedance analysis often fail to capture nonlinear, high-dimensional biomedical relationships. We propose a fuzzy neuro-evolutionary multi-objective framework that optimizes feature subsets for Multi-Layer Perceptron (MLP) regression and selects clinically balanced trade-offs using a Nash-bargaining-based decision layer. Experiments on the Johnson anthropometric dataset (n = 252) and the NHANES 1999–2000 clinical dataset (n = 862) show that the single- and multi-objective variants reduce feature dimensionality by up to 65 % relative to the full feature set while achieving a 3–8 % reduction in RMSE compared with state-of-the-art fuzzy and evolutionary baselines. The proposed approach also improves stability, reflected by lower within-run residual dispersion (reported as the run-averaged value <span><math><msub><mrow><mi>STD</mi></mrow><mrow><mi>err</mi><mo>,</mo><mi>mean</mi></mrow></msub></math></span> and lower run-to-run variability (<span><math><mrow><msub><mrow><mi>STD</mi></mrow><mrow><mi>runs</mi></mrow></msub><mo>(</mo><mo>.</mo><mo>)</mo></mrow></math></span>) over 30 independent runs; Wilcoxon signed-rank tests confirm statistically significant improvements (p < 0.05). The multi-objective model jointly optimizes parsimony (<span><math><msub><mrow><mi>n</mi></mrow><mrow><mi>f</mi></mrow></msub></math></span>), prediction accuracy (RMSE), and robustness measured by within-run residual dispersion (<span><math><msub><mrow><mi>STD</mi></mrow><mrow><mi>err</mi></mrow></msub></math></span>), and selects fair Pareto solutions via a cooperative Multi-Objective Bargaining Game. A symmetric V-shaped transfer function empirically improves binary neuro-evolution stability. The resulting models provide interpretable and clinically plausible predictors for body-fat estimation, supporting trustworthy biomedical decision support.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"675 ","pages":"Article 132974"},"PeriodicalIF":6.5,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172923","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

TFIGF: Fire data augmentation model based on text-to-image synthesis TFIGF：基于文本图像合成的火灾数据增强模型

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-05 DOI: 10.1016/j.neucom.2026.132912

Hongyang Zhao , Yanan Guo , Xingdong Li , Yi Liu , Jing Jin

Data imbalance is one of the most challenging issues in deep learning, particularly in the domain of fire detection. In this field, the number of non-fire images significantly exceeds that of fire images, and the diversity of background information in images poses substantial challenges to fire detection. Recently, there have been significant advancements in generating images from textual descriptions using large language models. Inspired by this progress, this paper proposes an innovative Text-to-Image Fire Image Generation Framework (TFIGF). This framework aims to address the problem of insufficient model training due to a lack of adequate positive samples by generating fire images with varied backgrounds, thereby enhancing the efficiency and accuracy of fire detection. The proposed TFIGF framework consists of a front-end image generator and a back-end image filter. The image generator, comprising a feature fusion component, a CLIP image encoder based on the Vision Transformer (ViT), and a feature generation segment, is capable of merging textual information with the prior knowledge in the pre-trained CLIP-ViT model to produce images, enhancing the relevance and diversity of the generated images. Images produced by the image generator are evaluated and filtered by the image filter to obtain fire images most congruent with the textual descriptions. The proposed image filter converts the generated visual information into textual descriptions using ViT and GPT-3, and measures the alignment between the generated images and input text using cosine similarity. The proposed method can generate higher-quality images compared to state-of-the-art generative image methods. Furthermore, to verify the improvements in accuracy and reliability of fire detection with images generated by TFIGF, we constructed datasets augmented to various sizes, trained several popular detection models on these datasets, and tested them with real-world data. Experimental results demonstrate that images generated by TFIGF significantly enhance network performance in fire detection, confirming the framework’s potential and practicality in addressing data imbalance issues.

数据不平衡是深度学习中最具挑战性的问题之一，特别是在火灾探测领域。在该领域，非火灾图像的数量大大超过了火灾图像的数量，并且图像中背景信息的多样性给火灾检测带来了很大的挑战。最近，在使用大型语言模型从文本描述生成图像方面取得了重大进展。受此启发，本文提出了一种创新的文本到图像火焰图像生成框架（TFIGF）。该框架旨在通过生成不同背景的火灾图像，解决由于缺乏足够的正样本而导致的模型训练不足的问题，从而提高火灾探测的效率和准确性。提出的TFIGF框架由前端图像生成器和后端图像滤波器组成。该图像生成器包括特征融合组件、基于视觉变换（Vision Transformer, ViT）的CLIP图像编码器和特征生成片段，能够将文本信息与预训练的CLIP-ViT模型中的先验知识合并生成图像，增强生成图像的相关性和多样性。图像生成器生成的图像由图像过滤器进行评估和过滤，以获得与文本描述最一致的5张图像。该图像滤波器利用ViT和GPT-3将生成的视觉信息转换为文本描述，并利用余弦相似度度量生成的图像与输入文本的对齐程度。与最先进的生成图像方法相比，所提出的方法可以生成更高质量的图像。此外，为了验证使用TFIGF生成的图像进行火灾探测的准确性和可靠性的提高，我们构建了扩充到不同大小的数据集，在这些数据集上训练了几种流行的探测模型，并用实际数据对它们进行了测试。实验结果表明，TFIGF生成的图像显著提高了网络在火灾检测中的性能，证实了该框架在解决数据不平衡问题方面的潜力和实用性。

{"title":"TFIGF: Fire data augmentation model based on text-to-image synthesis","authors":"Hongyang Zhao , Yanan Guo , Xingdong Li , Yi Liu , Jing Jin","doi":"10.1016/j.neucom.2026.132912","DOIUrl":"10.1016/j.neucom.2026.132912","url":null,"abstract":"<div><div>Data imbalance is one of the most challenging issues in deep learning, particularly in the domain of fire detection. In this field, the number of non-fire images significantly exceeds that of fire images, and the diversity of background information in images poses substantial challenges to fire detection. Recently, there have been significant advancements in generating images from textual descriptions using large language models. Inspired by this progress, this paper proposes an innovative Text-to-Image Fire Image Generation Framework (TFIGF). This framework aims to address the problem of insufficient model training due to a lack of adequate positive samples by generating fire images with varied backgrounds, thereby enhancing the efficiency and accuracy of fire detection. The proposed TFIGF framework consists of a front-end image generator and a back-end image filter. The image generator, comprising a feature fusion component, a CLIP image encoder based on the Vision Transformer (ViT), and a feature generation segment, is capable of merging textual information with the prior knowledge in the pre-trained CLIP-ViT model to produce images, enhancing the relevance and diversity of the generated images. Images produced by the image generator are evaluated and filtered by the image filter to obtain fire images most congruent with the textual descriptions. The proposed image filter converts the generated visual information into textual descriptions using ViT and GPT-3, and measures the alignment between the generated images and input text using cosine similarity. The proposed method can generate higher-quality images compared to state-of-the-art generative image methods. Furthermore, to verify the improvements in accuracy and reliability of fire detection with images generated by TFIGF, we constructed datasets augmented to various sizes, trained several popular detection models on these datasets, and tested them with real-world data. Experimental results demonstrate that images generated by TFIGF significantly enhance network performance in fire detection, confirming the framework’s potential and practicality in addressing data imbalance issues.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"675 ","pages":"Article 132912"},"PeriodicalIF":6.5,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146147216","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

Robust jointly sparse 2-dimensional projection fuzzy clustering with local manifold structure preservation 具有局部流形结构保留的鲁棒联合稀疏二维投影模糊聚类

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-04 DOI: 10.1016/j.neucom.2026.132970

Wu Chengmao , Fengchao Gong

Dimensionality reduction clustering methods combine feature reduction and clustering to analyze high-dimensional image data. However, 1D projection subspace clustering vectorizes 2D images into 1D vectors, disrupting spatial correlations and causing information loss. Two-stage models that separate reduction and clustering lack coordination, leading to suboptimal results. We propose a robust sparse two-dimensional projection fuzzy clustering method with local manifold constraints to improve image clustering. Each cluster is represented by a bilinear orthogonal subspace, and F1-norm reconstruction error updates sample memberships. A similarity matrix captures affinities, while a Laplacian matrix preserves manifold geometry during dimensionality reduction. Optimization uses block coordinate descent to alternately refine the projection matrix, cluster centroids, and membership matrix until convergence. This unified, unsupervised model avoids image vectorization, reducing computational complexity and preserving spatial relationships. Experiments on nine benchmark datasets show the RS2DPFC-LMS algorithm improves accuracy by 2.47 % and normalized mutual information by 2 %, demonstrating superior clustering performance, parameter stability, and noise robustness.

降维聚类方法将特征约简和聚类相结合，对高维图像数据进行分析。然而，一维投影子空间聚类将二维图像矢量化为一维向量，破坏了空间相关性，造成信息丢失。分离约简和聚类的两阶段模型缺乏协调，导致次优结果。提出了一种具有局部流形约束的鲁棒稀疏二维投影模糊聚类方法。每个聚类由双线性正交子空间表示，f1范数重构误差更新样本隶属度。相似矩阵捕获相似性，而拉普拉斯矩阵在降维过程中保留流形几何。优化采用分块坐标下降交替优化投影矩阵、聚类质心和隶属度矩阵，直至收敛。这种统一的无监督模型避免了图像矢量化，降低了计算复杂度并保留了空间关系。在9个基准数据集上的实验表明，RS2DPFC-LMS算法的准确率提高了2.47 %，归一化互信息提高了2 %，表现出了优异的聚类性能、参数稳定性和噪声鲁棒性。

{"title":"Robust jointly sparse 2-dimensional projection fuzzy clustering with local manifold structure preservation","authors":"Wu Chengmao , Fengchao Gong","doi":"10.1016/j.neucom.2026.132970","DOIUrl":"10.1016/j.neucom.2026.132970","url":null,"abstract":"<div><div>Dimensionality reduction clustering methods combine feature reduction and clustering to analyze high-dimensional image data. However, 1D projection subspace clustering vectorizes 2D images into 1D vectors, disrupting spatial correlations and causing information loss. Two-stage models that separate reduction and clustering lack coordination, leading to suboptimal results. We propose a robust sparse two-dimensional projection fuzzy clustering method with local manifold constraints to improve image clustering. Each cluster is represented by a bilinear orthogonal subspace, and F1-norm reconstruction error updates sample memberships. A similarity matrix captures affinities, while a Laplacian matrix preserves manifold geometry during dimensionality reduction. Optimization uses block coordinate descent to alternately refine the projection matrix, cluster centroids, and membership matrix until convergence. This unified, unsupervised model avoids image vectorization, reducing computational complexity and preserving spatial relationships. Experiments on nine benchmark datasets show the RS2DPFC-LMS algorithm improves accuracy by 2.47 % and normalized mutual information by 2 %, demonstrating superior clustering performance, parameter stability, and noise robustness.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"676 ","pages":"Article 132970"},"PeriodicalIF":6.5,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173297","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

FedACA: Adaptive classifier aggregation and clustering for personalized heterogeneous federated learning FedACA：用于个性化异构联邦学习的自适应分类器聚合和聚类

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-04 DOI: 10.1016/j.neucom.2026.132972

Jichen Dong , Yingchun Cui , Zhengda Wu , Heran Xi , Jinghua Zhu

Heterogeneous federated learning aims to address the challenges of slow convergence and high communication cost caused by data, model, and device heterogeneity. Prototype based approaches reduce communication costs through sharing class prototypes instead of model parameters across clients. However, existing methods still suffer from prototype shift, boundary ambiguity, and distribution skew. To overcome these challenges, we propose FedACA(Federated Adaptive Clustered Aggregation), a new federated learning framework with adaptive clustering and personalized classifier aggregation. FedACA has three core innovations: Adaptive K-means clustering, which can improve the quality of local prototypes through adaptive cluster center optimization, instead of simple feature averaging; Cluster head matching, which clusters the clients with similar prototypes into a group and fuses their local classifiers into a hybrid classifier to mitigate the class imbalance problem; Personalized head fusion, in which clients adaptively fuse global and local classifiers, refining fine-grained decision boundaries through meta-learning-guided aggregation. We conduct extensive experiments on various datasets and compare our FedACA with the existing methods. The results show that our method achieves 9.66 % accuracy improvement in heterogeneous settings, which demonstrates the effectiveness of our method under diverse conditions.

异构联邦学习旨在解决由数据、模型和设备异构引起的缓慢收敛和高通信成本的挑战。基于原型的方法通过在客户端之间共享类原型而不是模型参数来降低通信成本。然而，现有的方法仍然存在原型偏移、边界模糊和分布不均匀等问题。为了克服这些挑战，我们提出了FedACA（联邦自适应聚类聚合），这是一种新的联邦学习框架，具有自适应聚类和个性化分类器聚合。FedACA有三个核心创新：自适应K-means聚类，通过自适应聚类中心优化来提高局部原型的质量，而不是简单的特征平均；簇头匹配，将具有相似原型的客户端聚为一组，并将其局部分类器融合成混合分类器，以缓解类不平衡问题；个性化头部融合，其中客户端自适应融合全局和局部分类器，通过元学习引导的聚合来细化细粒度的决策边界。我们在各种数据集上进行了大量的实验，并将我们的FedACA与现有方法进行了比较。结果表明，该方法在异构环境下的准确率提高了9.66%，证明了该方法在不同条件下的有效性。

{"title":"FedACA: Adaptive classifier aggregation and clustering for personalized heterogeneous federated learning","authors":"Jichen Dong , Yingchun Cui , Zhengda Wu , Heran Xi , Jinghua Zhu","doi":"10.1016/j.neucom.2026.132972","DOIUrl":"10.1016/j.neucom.2026.132972","url":null,"abstract":"<div><div>Heterogeneous federated learning aims to address the challenges of slow convergence and high communication cost caused by data, model, and device heterogeneity. Prototype based approaches reduce communication costs through sharing class prototypes instead of model parameters across clients. However, existing methods still suffer from prototype shift, boundary ambiguity, and distribution skew. To overcome these challenges, we propose FedACA(Federated Adaptive Clustered Aggregation), a new federated learning framework with adaptive clustering and personalized classifier aggregation. FedACA has three core innovations: <strong>Adaptive K-means clustering</strong>, which can improve the quality of local prototypes through adaptive cluster center optimization, instead of simple feature averaging; <strong>Cluster head matching</strong>, which clusters the clients with similar prototypes into a group and fuses their local classifiers into a hybrid classifier to mitigate the class imbalance problem; <strong>Personalized head fusion</strong>, in which clients adaptively fuse global and local classifiers, refining fine-grained decision boundaries through meta-learning-guided aggregation. We conduct extensive experiments on various datasets and compare our FedACA with the existing methods. The results show that our method achieves 9.66 % accuracy improvement in heterogeneous settings, which demonstrates the effectiveness of our method under diverse conditions.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"674 ","pages":"Article 132972"},"PeriodicalIF":6.5,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146192243","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

Weightless multi-class classifier ruled by “Mental images” 由“心理意象”支配的失重多类分类器

IF 6.5 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neurocomputing

Pub Date : 2026-02-04 DOI: 10.1016/j.neucom.2026.132975

Antonio Sorgente, Massimo De Gregorio

Common-sense rules represent a form of implicit knowledge derived from experience and observation, often used by both humans and artificial systems to guide decision-making. In this work, we introduce a novel extension of the DRASiW weightless neural network architecture that enables the automatic extraction of such rules from its internal knowledge representations, known as “Mental Images” (MIs). These rules, grounded in statistical patterns learned during training, are used to selectively inhibit a subset of discriminators prior to classification, effectively acting as a pre-filtering mechanism. The inhibition process is designed to reduce the computational burden and improve classification plausibility by narrowing the focus to the most promising classes. Unlike previous approaches that use MIs solely for interpretation or symbolic conversion, our method exploits this implicit knowledge to enhance the system’s own performance. The extracted rules reflect structural regularities within the input space and serve as a proxy for common sense reasoning embedded within the trained model.

This work highlights how internal representations in weightless neural systems can be explicitly leveraged to support interpretable and efficient decision processes, bridging the gap between subsymbolic learning and symbolic reasoning through rule-based inhibition.

常识性规则代表了一种从经验和观察中获得的隐性知识，通常被人类和人工系统用来指导决策。在这项工作中，我们引入了DRASiW无权重神经网络架构的新扩展，该架构能够从其内部知识表示（称为“心理图像”（MIs））中自动提取这些规则。这些规则基于训练期间学习到的统计模式，用于在分类之前选择性地抑制鉴别器子集，有效地充当预过滤机制。抑制过程旨在通过将焦点缩小到最有希望的类别来减少计算负担并提高分类的合理性。不像以前的方法只使用MIs进行解释或符号转换，我们的方法利用这种隐性知识来提高系统自身的性能。提取的规则反映了输入空间中的结构规律，并作为嵌入在训练模型中的常识推理的代理。这项工作强调了如何明确利用失重神经系统中的内部表征来支持可解释和有效的决策过程，通过基于规则的抑制弥合亚符号学习和符号推理之间的差距。

{"title":"Weightless multi-class classifier ruled by “Mental images”","authors":"Antonio Sorgente, Massimo De Gregorio","doi":"10.1016/j.neucom.2026.132975","DOIUrl":"10.1016/j.neucom.2026.132975","url":null,"abstract":"<div><div>Common-sense rules represent a form of implicit knowledge derived from experience and observation, often used by both humans and artificial systems to guide decision-making. In this work, we introduce a novel extension of the DRASiW weightless neural network architecture that enables the automatic extraction of such rules from its internal knowledge representations, known as “Mental Images” (MIs). These rules, grounded in statistical patterns learned during training, are used to selectively inhibit a subset of discriminators prior to classification, effectively acting as a pre-filtering mechanism. The inhibition process is designed to reduce the computational burden and improve classification plausibility by narrowing the focus to the most promising classes. Unlike previous approaches that use MIs solely for interpretation or symbolic conversion, our method exploits this implicit knowledge to enhance the system’s own performance. The extracted rules reflect structural regularities within the input space and serve as a proxy for common sense reasoning embedded within the trained model.</div><div>This work highlights how internal representations in weightless neural systems can be explicitly leveraged to support interpretable and efficient decision processes, bridging the gap between subsymbolic learning and symbolic reasoning through rule-based inhibition.</div></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":"675 ","pages":"Article 132975"},"PeriodicalIF":6.5,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146147422","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