Transfer_Learning
-
Coreset selection based on Intra-class diversity
2025-09-23 12:45:53Imran Ashraf, Mukhtar Ullah, Muhammad Faisal Nadeem, Muhammad Nouman NoorAbstract
Deep Learning models have transformed various domains, including the healthcare sector, particularly biomedical image classification by learning intricate features and enabling accurate diagnostics pertaining to complex diseases. Recent studies have adopted two different approaches to train DL models: training from scratch and transfer learning. Both approaches demand substantial computational time and resources due to the involvement of massive datasets in model training. These computational demands are further increased due to the design-space exploration required for selecting optimal hyperparameters, which typically necessitates several training rounds. With the growing sizes of datasets, exploring solutions to this problem has recently gained the research community's attention. A plausible solution is to select a subset of the dataset for training and hyperparameter search. This subset, referred to as the corset, must be a representative set of the original dataset. A straightforward approach to selecting the coreset could be employing random sampling, albeit at the cost of compromising the representativeness of the original dataset. A critical limitation of random sampling is the bias towards the dominant classes in an imbalanced dataset. Even if the dataset has inter-class balance, this random sampling will not capture intra-class diversity. This study addresses this issue by introducing an intelligent, lightweight mechanism for coreset selection. Specifically, it proposes a method to extract intra-class diversity, forming per-class clusters that are utilized for the final sampling. We demonstrate the efficacy of the proposed methodology by conducting extensive classification experiments on a well-known biomedical imaging dataset. Results demonstrate that the proposed scheme outperforms the random sampling approach on several performance metrics for uniform conditions.
Abstract (translated)
深度学习模型已经变革了包括医疗保健在内的各个领域，尤其是在通过学习复杂特征并使复杂疾病的准确诊断成为可能的生物医学图像分类方面。最近的研究采用了两种不同的方法来训练深度学习（DL）模型：从头开始训练和迁移学习。这两种方法都需要大量的计算时间和资源，因为大规模数据集在模型训练中起着关键作用。由于需要进行设计空间探索以选择最优超参数，通常需要多次训练轮次，这进一步增加了计算需求。随着数据集规模的增加，解决这一问题的方法越来越受到研究社区的关注。一种可能的解决方案是从整个数据集中选取一个子集用于训练和超参数搜索。这个子集被称为“核心集”（coreset），必须代表原始数据集。简单地通过随机抽样来选择核心集的做法虽然可以实现，但可能会牺牲原始数据集的代表性。随机采样的一个重要限制是对不平衡数据集中主导类别的偏见。即使在类别间平衡的数据集中，随机采样也无法捕捉到类内的多样性。这项研究通过引入一种智能且轻量级的核心集选取机制来解决这一问题。具体而言，它提出了一种提取类内多样性的方法，形成用于最终采样的每类聚类。我们通过对一个著名的生物医学成像数据集进行广泛的分类实验，展示了所提方法的有效性。结果表明，在统一条件下，该方案在多个性能指标上优于随机抽样方法。
URL
https://arxiv.org/abs/2509.21380
PDF
-
Efficient Breast and Ovarian Cancer Classification via ViT-Based Preprocessing and Transfer Learning
2025-09-23 02:25:44Richa Rawat, Faisal AhmedAbstract
Cancer is one of the leading health challenges for women, specifically breast and ovarian cancer. Early detection can help improve the survival rate through timely intervention and treatment. Traditional methods of detecting cancer involve manually examining mammograms, CT scans, ultrasounds, and other imaging types. However, this makes the process labor-intensive and requires the expertise of trained pathologists. Hence, making it both time-consuming and resource-intensive. In this paper, we introduce a novel vision transformer (ViT)-based method for detecting and classifying breast and ovarian cancer. We use a pre-trained ViT-Base-Patch16-224 model, which is fine-tuned for both binary and multi-class classification tasks using publicly available histopathological image datasets. Further, we use a preprocessing pipeline that converts raw histophological images into standardized PyTorch tensors, which are compatible with the ViT architecture and also help improve the model performance. We evaluated the performance of our model on two benchmark datasets: the BreakHis dataset for binary classification and the UBC-OCEAN dataset for five-class classification without any data augmentation. Our model surpasses existing CNN, ViT, and topological data analysis-based approaches in binary classification. For multi-class classification, it is evaluated against recent topological methods and demonstrates superior performance. Our study highlights the effectiveness of Vision Transformer-based transfer learning combined with efficient preprocessing in oncological diagnostics.
Abstract (translated)
癌症是女性面临的重大健康挑战之一，特别是乳腺癌和卵巢癌。早期检测有助于通过及时的干预和治疗来提高生存率。传统上，癌症检测方法包括手动检查乳房X线摄影（钼靶）、CT扫描、超声波等影像资料。然而，这使得过程变得繁琐，并且需要有经验的专业病理学家进行解读，从而耗时且耗费资源。 在本文中，我们提出了一种基于视觉变换器（Vision Transformer, ViT）的新方法来检测和分类乳腺癌与卵巢癌。我们使用了一个预训练的ViT-Base-Patch16-224模型，并针对二元分类和多类分类任务，在公开的历史病理图像数据集上进行了微调。此外，我们还设计了一套预处理管道，将原始历史病理图片转化为标准化的PyTorch张量，这些张量与ViT架构兼容并有助于提升模型性能。 我们在两个基准数据集上评估了我们的模型的表现：BreakHis数据集用于二元分类任务，UBC-OCEAN数据集用于五类无数据增强情况下的多类别分类。在二元分类中，我们的模型超越了现有的卷积神经网络（CNN）、ViT和基于拓扑数据分析的方法；而在多类分类方面，它与最新的拓扑方法进行了对比，并展示了更为优越的性能。 本研究强调了视觉变换器基础的迁移学习与高效的预处理结合在肿瘤诊断中的有效性。
URL
https://arxiv.org/abs/2509.18553
PDF
-
TS-P$^2$CL: Plug-and-Play Dual Contrastive Learning for Vision-Guided Medical Time Series Classification
2025-09-22 13:57:58Qi'ao Xu, Pengfei Wang, Bo Zhong, Tianwen Qian, Xiaoling Wang, Ye Wang, Hong YuAbstract
Medical time series (MedTS) classification is pivotal for intelligent healthcare, yet its efficacy is severely limited by poor cross-subject generation due to the profound cross-individual heterogeneity. Despite advances in architectural innovations and transfer learning techniques, current methods remain constrained by modality-specific inductive biases that limit their ability to learn universally invariant representations. To overcome this, we propose TS-P$^2$CL, a novel plug-and-play framework that leverages the universal pattern recognition capabilities of pre-trained vision models. We introduce a vision-guided paradigm that transforms 1D physiological signals into 2D pseudo-images, establishing a bridge to the visual domain. This transformation enables implicit access to rich semantic priors learned from natural images. Within this unified space, we employ a dual-contrastive learning strategy: intra-modal consistency enforces temporal coherence, while cross-modal alignment aligns time-series dynamics with visual semantics, thereby mitigating individual-specific biases and learning robust, domain-invariant features. Extensive experiments on six MedTS datasets demonstrate that TS-P$^2$CL consistently outperforms fourteen methods in both subject-dependent and subject-independent settings.
Abstract (translated)
医学时间序列（MedTS）分类在智能医疗领域至关重要，但其效果因个体间显著的异质性导致跨主体泛化能力差而受到严重限制。尽管在架构创新和迁移学习技术方面取得了进展，目前的方法仍然受制于模式特定的归纳偏差，这限制了它们学习普遍不变特征的能力。为解决这一问题，我们提出了TS-P$^2$CL，这是一种新颖的即插即用框架，利用预训练视觉模型的通用模式识别能力。我们引入了一种以视觉为导向的范式，将1D生理信号转换成2D伪图像，在此基础上建立了一个通往视觉领域的桥梁。这种转换使隐性访问从自然图像中学习到的丰富语义先验成为可能。在这个统一的空间内，我们采用了双对比学习策略：模式内部一致性强制时间连贯性，而跨模态对齐将时间序列动态与视觉语义对齐，从而减轻个体特异性偏差，并学习稳健、域不变的特征。在六个MedTS数据集上的广泛实验表明，在依赖于主体和不依赖于主体的情况下，TS-P$^2$CL方法始终优于其他十四种方法。
URL
https://arxiv.org/abs/2509.17802
PDF
-
Enhancing Cluster Scheduling in HPC: A Continuous Transfer Learning for Real-Time Optimization
2025-09-22 12:27:20Leszek Sliwko, Jolanta Mizera-PietraszkoAbstract
This study presents a machine learning-assisted approach to optimize task scheduling in cluster systems, focusing on node-affinity constraints. Traditional schedulers like Kubernetes struggle with real-time adaptability, whereas the proposed continuous transfer learning model evolves dynamically during operations, minimizing retraining needs. Evaluated on Google Cluster Data, the model achieves over 99% accuracy, reducing computational overhead and improving scheduling latency for constrained tasks. This scalable solution enables real-time optimization, advancing machine learning integration in cluster management and paving the way for future adaptive scheduling strategies.
Abstract (translated)
这项研究提出了一种利用机器学习辅助优化集群系统任务调度的方法，重点在于节点亲和性约束。传统的调度器如Kubernetes在实时适应性方面存在困难，而所提出的连续迁移学习模型则可以在运行过程中动态演变，从而减少重新训练的需求。通过谷歌集群数据的评估，该模型实现了超过99%的准确率，减少了计算开销，并提高了受限制任务的调度延迟。这一可扩展的解决方案使实时优化成为可能，推进了机器学习在集群管理中的集成，并为未来的自适应调度策略铺平了道路。
URL
https://arxiv.org/abs/2509.22701
PDF
-
CUTE: A Multilingual Dataset for Enhancing Cross-Lingual Knowledge Transfer in Low-Resource Languages
2025-09-21 04:30:49Wenhao Zhuang, Yuan SunAbstract
Large Language Models (LLMs) demonstrate exceptional zero-shot capabilities in various NLP tasks, significantly enhancing user experience and efficiency. However, this advantage is primarily limited to resource-rich languages. For the diverse array of low-resource languages, support remains inadequate, with the scarcity of training corpora considered the primary cause. We construct and open-source CUTE Chinese, Uyghur, Tibetan,English dataset, consisting of two 25GB sets of four-language corpora (one parallel and one non-parallel), obtained through machine translation. CUTE encompasses two resource-rich languages (Chinese and English) and two low-resource languages (Uyghur and Tibetan). Prior to constructing CUTE, human assessment validates that the machine translation quality between Chinese-Uyghur and Chinese-Tibetan approaches that of Chinese-English translation. CUTE represents the largest open-source corpus for Uyghur and Tibetan languages to date, and we demonstrate its effectiveness in enhancing LLMs' ability to process low-resource languages while investigating the role of corpus parallelism in cross-lingual transfer learning. The CUTE corpus and related models are made publicly available to the research community.
Abstract (translated)
大型语言模型（LLMs）在各种自然语言处理任务中表现出卓越的零样本学习能力，显著提升了用户体验和效率。然而，这一优势主要局限于资源丰富的语言。对于众多低资源语言而言，支持仍然不足，训练语料库的稀缺被认为是主要原因。 我们构建并开源了CUTE（Chinese, Uyghur, Tibetan, English）数据集，该数据集包含两组共计25GB的四语言语料库（一组为平行语料，另一组为非平行语料），这些语料通过机器翻译获得。CUTE涵盖了两种资源丰富的语言（中文和英文）以及两种低资源语言（维吾尔语和藏语）。在构建CUTE之前，经人工评估确认，从汉语到维吾尔语和从汉语到藏语的机器翻译质量接近于汉语到英语的翻译质量。 CUTE是迄今为止最大的公开可用的维吾尔语和藏语文本语料库。我们展示了该数据集能够增强LLMs处理低资源语言的能力，并研究了平行语料在跨语言迁移学习中的作用。CUTE语料及其相关模型已向研究界公开发布。
URL
https://arxiv.org/abs/2509.16914
PDF
-
Attention Schema-based Attention Control : A Cognitive-Inspired Approach for Attention Management in Transformers
2025-09-19 15:08:30Krati Saxena, Federico Jurado Ruiz, Guido Manzi, Dianbo Liu, Alex LambAbstract
Attention mechanisms have become integral in AI, significantly enhancing model performance and scalability by drawing inspiration from human cognition. Concurrently, the Attention Schema Theory (AST) in cognitive science posits that individuals manage their attention by creating a model of the attention itself, effectively allocating cognitive resources. Inspired by AST, we introduce ASAC (Attention Schema-based Attention Control), which integrates the attention schema concept into artificial neural networks. Our initial experiments focused on embedding the ASAC module within transformer architectures. This module employs a Vector-Quantized Variational AutoEncoder (VQVAE) as both an attention abstractor and controller, facilitating precise attention management. By explicitly modeling attention allocation, our approach aims to enhance system efficiency. We demonstrate ASAC's effectiveness in both the vision and NLP domains, highlighting its ability to improve classification accuracy and expedite the learning process. Our experiments with vision transformers across various datasets illustrate that the attention controller not only boosts classification accuracy but also accelerates learning. Furthermore, we have demonstrated the model's robustness and generalization capabilities across noisy and out-of-distribution datasets. In addition, we have showcased improved performance in multi-task settings. Quick experiments reveal that the attention schema-based module enhances resilience to adversarial attacks, optimizes attention to improve learning efficiency, and facilitates effective transfer learning and learning from fewer examples. These promising results establish a connection between cognitive science and machine learning, shedding light on the efficient utilization of attention mechanisms in AI systems.
Abstract (translated)
注意力机制在人工智能领域中变得至关重要，通过借鉴人类认知的灵感显著提高了模型的表现和可扩展性。同时，在认知科学中提出的注意模式理论（AST）认为个体通过创建注意力的模型来管理自己的注意力，从而有效地分配认知资源。受AST启发，我们引入了基于注意模式的注意力控制（ASAC），将注意模式概念融入人工神经网络之中。我们的初步实验重点在于在变压器架构内嵌入ASAC模块。该模块采用向量量化变分自动编码器（VQVAE）作为注意力抽象器和控制器，以便实现精确的注意力管理。通过显式地建模注意力分配，我们旨在提高系统的效率。 我们在视觉和自然语言处理（NLP）领域展示了ASAC的有效性，突出了其提高分类准确性和加快学习过程的能力。在不同数据集上的视觉变压器实验表明，注意控制器不仅提高了分类准确性，还加速了学习速度。此外，我们已经证明该模型在噪声和分布外数据集上具有鲁棒性和泛化能力。另外，在多任务设置中也展示了改进的性能。 快速实验揭示了基于注意力模式的模块增强了对对抗攻击的抵抗力、优化了注意力以提高学习效率，并促进了有效的迁移学习以及从少量样本中的学习。这些有前景的结果建立了认知科学和机器学习之间的联系，阐明了在AI系统中有效利用注意机制的方法。
URL
https://arxiv.org/abs/2509.16058
PDF
-
Frequency-Aware Ensemble Learning for BraTS 2025 Pediatric Brain Tumor Segmentation
2025-09-18 03:24:58Yuxiao Yi, Qingyao Zhuang, Zhi-Qin John XuAbstract
Pediatric brain tumor segmentation presents unique challenges due to the rarity and heterogeneity of these malignancies, yet remains critical for clinical diagnosis and treatment planning. We propose an ensemble approach integrating nnU-Net, Swin UNETR, and HFF-Net for the BraTS-PED 2025 challenge. Our method incorporates three key extensions: adjustable initialization scales for optimal nnU-Net complexity control, transfer learning from BraTS 2021 pre-trained models to enhance Swin UNETR's generalization on pediatric dataset, and frequency domain decomposition for HFF-Net to separate low-frequency tissue contours from high-frequency texture details. Our final ensemble combines nnU-Net ($\gamma=0.7$), fine-tuned Swin UNETR, and HFF-Net, achieving Dice scores of 72.3% (ET), 95.6% (NET), 68.9% (CC), 89.5% (ED), 92.3% (TC), and 92.3% (WT), respectively.
Abstract (translated)
儿童脑肿瘤分割由于这些恶性肿瘤的罕见性和异质性而面临独特的挑战，但它对于临床诊断和治疗计划仍然至关重要。我们为BraTS-PED 2025挑战赛提出了一种集成方法，结合了nnU-Net、Swin UNETR和HFF-Net。我们的方法包含了三个关键扩展：可调初始化尺度以实现最佳的nnU-Net复杂度控制；从BraTS 2021预训练模型中进行迁移学习以增强Swin UNETR在儿童数据集上的泛化能力；以及通过频率域分解使HFF-Net能够将低频组织轮廓与高频纹理细节分离。最终集成方法包括调整参数为$\gamma=0.7$的nnU-Net、微调后的Swin UNETR和HFF-Net，分别取得了72.3%（增强肿瘤, ET）、95.6%（非增强肿瘤, NET）、68.9%（核心肿瘤, CC）、89.5%（全脑水肿, ED）、92.3%（全肿瘤, TC）和92.3%（全病变, WT）的Dice评分。
URL
https://arxiv.org/abs/2509.19353
PDF
-
R-Net: A Reliable and Resource-Efficient CNN for Colorectal Cancer Detection with XAI Integration
2025-09-17 18:29:44Rokonozzaman Ayon, Md Taimur Ahad, Bo Song, Yan LiAbstract
State-of-the-art (SOTA) Convolutional Neural Networks (CNNs) are criticized for their extensive computational power, long training times, and large datasets. To overcome this limitation, we propose a reasonable network (R-Net), a lightweight CNN only to detect and classify colorectal cancer (CRC) using the Enteroscope Biopsy Histopathological Hematoxylin and Eosin Image Dataset (EBHI). Furthermore, six SOTA CNNs, including Multipath-based CNNs (DenseNet121, ResNet50), Depth-based CNNs (InceptionV3), width-based multi-connection CNNs (Xception), depth-wise separable convolutions (MobileNetV2), spatial exploitation-based CNNs (VGG16), Transfer learning, and two ensemble models are also tested on the same dataset. The ensemble models are a multipath-depth-width combination (DenseNet121-InceptionV3-Xception) and a multipath-depth-spatial combination (ResNet18-InceptionV3-VGG16). However, the proposed R-Net lightweight achieved 99.37% accuracy, outperforming MobileNet (95.83%) and ResNet50 (96.94%). Most importantly, to understand the decision-making of R-Net, Explainable AI such as SHAP, LIME, and Grad-CAM are integrated to visualize which parts of the EBHI image contribute to the detection and classification process of R-Net. The main novelty of this research lies in building a reliable, lightweight CNN R-Net that requires fewer computing resources yet maintains strong prediction results. SOTA CNNs, transfer learning, and ensemble models also extend our knowledge on CRC classification and detection. XAI functionality and the impact of pixel intensity on correct and incorrect classification images are also some novelties in CRC detection and classification.
Abstract (translated)
最先进的（SOTA）卷积神经网络（CNN）因其计算能力要求高、训练时间长和需要大量数据而受到批评。为克服这一限制，我们提出了一种合理的网络（R-Net），这是一种轻量级的CNN，用于使用肠镜活检组织病理学苏木精和伊红图像数据集（EBHI）检测和分类结直肠癌（CRC）。此外，还测试了六个SOTA CNN模型，包括基于多路径的CNNs（DenseNet121、ResNet50）、基于深度的CNNs（InceptionV3）、基于宽度的多连接CNNs（Xception）、深度可分离卷积（MobileNetV2）、空间探索型CNNs（VGG16）以及迁移学习和两个集成模型。这些集成模型分别是多路径-深度-宽度组合（DenseNet121-InceptionV3-Xception）和多路径-深度-空间组合（ResNet18-InceptionV3-VGG16）。然而，提出的轻量级R-Net达到了99.37%的准确率，超过了MobileNet（95.83%）和ResNet50（96.94%）。尤为重要的是，为了理解R-Net的决策过程，我们集成了可解释的人工智能技术如SHAP、LIME和Grad-CAM来可视化EBHI图像中哪些部分对R-Net的检测和分类过程有贡献。 这项研究的主要创新之处在于构建了一个可靠且轻量级的CNN R-Net，它需要较少的计算资源但保持了强大的预测性能。SOTA CNNs、迁移学习以及集成模型也扩展了我们对CRC分类和检测的理解。XAI功能（如解释性人工智能）以及像素强度对正确与错误分类图像的影响也是CRC检测和分类的一些新特点。
URL
https://arxiv.org/abs/2509.16251
PDF
-
A study on Deep Convolutional Neural Networks, transfer learning, and Mnet model for Cervical Cancer Detection
2025-09-17 18:11:09Saifuddin Sagor, Md Taimur Ahad, Faruk Ahmed, Rokonozzaman Ayon, Sanzida ParvinAbstract
Early and accurate detection through Pap smear analysis is critical to improving patient outcomes and reducing mortality of Cervical cancer. State-of-the-art (SOTA) Convolutional Neural Networks (CNNs) require substantial computational resources, extended training time, and large datasets. In this study, a lightweight CNN model, S-Net (Simple Net), is developed specifically for cervical cancer detection and classification using Pap smear images to address these limitations. Alongside S-Net, six SOTA CNNs were evaluated using transfer learning, including multi-path (DenseNet201, ResNet152), depth-based (Serasnet152), width-based multi-connection (Xception), depth-wise separable convolutions (MobileNetV2), and spatial exploitation-based (VGG19). All models, including S-Net, achieved comparable accuracy, with S-Net reaching 99.99%. However, S-Net significantly outperforms the SOTA CNNs in terms of computational efficiency and inference time, making it a more practical choice for real-time and resource-constrained applications. A major limitation in CNN-based medical diagnosis remains the lack of transparency in the decision-making process. To address this, Explainable AI (XAI) techniques, such as SHAP, LIME, and Grad-CAM, were employed to visualize and interpret the key image regions influencing model predictions. The novelty of this study lies in the development of a highly accurate yet computationally lightweight model (S-Net) caPable of rapid inference while maintaining interpretability through XAI integration. Furthermore, this work analyzes the behavior of SOTA CNNs, investigates the effects of negative transfer learning on Pap smear images, and examines pixel intensity patterns in correctly and incorrectly classified samples.
Abstract (translated)
通过巴氏涂片分析的早期和准确检测对于改善宫颈癌患者的预后并降低其死亡率至关重要。最先进的（SOTA）卷积神经网络（CNNs）需要大量的计算资源、长时间的训练以及大规模的数据集。在这项研究中，开发了一种轻量级的CNN模型——S-Net（简易网），专门用于使用巴氏涂片图像进行宫颈癌的检测和分类，以此来解决这些问题。 除了S-Net外，还利用迁移学习评估了六种最先进的CNNs，包括多路径（DenseNet201、ResNet152）、基于深度（Seranet152）、基于宽度的多重连接（Xception）、逐层可分离卷积（MobileNetV2）和基于空间探索（VGG19）。所有模型，包括S-Net，在准确性方面都达到了相当高的水平，其中S-Net的准确率达到99.99%。然而，与最先进的CNNs相比，S-Net在计算效率和推理时间上表现出色，使其成为实时应用及资源受限环境中的更优选择。 基于CNN的医学诊断的一个主要限制在于决策过程缺乏透明度。为解决这一问题，采用了可解释的人工智能（XAI）技术，如SHAP、LIME和Grad-CAM，以可视化并解释影响模型预测的关键图像区域。本研究的独特之处在于开发了一种高精度且计算效率高的模型——S-Net，能够在保持通过XAI集成的可解释性的同时实现快速推理。此外，这项工作还分析了最先进的CNNs的行为、探讨了负向迁移学习对巴氏涂片图像的影响，并检查了正确分类和错误分类样本中的像素强度模式。 这段翻译详细介绍了研究背景、模型开发目标以及所采用的技术手段和创新点，强调了S-Net在宫颈癌检测领域的应用价值及其优势。
URL
https://arxiv.org/abs/2509.16250
PDF
-
Automated Triaging and Transfer Learning of Incident Learning Safety Reports Using Large Language Representational Models
2025-09-17 05:29:23Peter Beidler, Mark Nguyen, Kevin Lybarger, Ola Holmberg, Eric Ford, John KangAbstract
PURPOSE: Incident reports are an important tool for safety and quality improvement in healthcare, but manual review is time-consuming and requires subject matter expertise. Here we present a natural language processing (NLP) screening tool to detect high-severity incident reports in radiation oncology across two institutions. METHODS AND MATERIALS: We used two text datasets to train and evaluate our NLP models: 7,094 reports from our institution (Inst.), and 571 from IAEA SAFRON (SF), all of which had severity scores labeled by clinical content experts. We trained and evaluated two types of models: baseline support vector machines (SVM) and BlueBERT which is a large language model pretrained on PubMed abstracts and hospitalized patient data. We assessed for generalizability of our model in two ways. First, we evaluated models trained using Inst.-train on SF-test. Second, we trained a BlueBERT_TRANSFER model that was first fine-tuned on Inst.-train then on SF-train before testing on SF-test set. To further analyze model performance, we also examined a subset of 59 reports from our Inst. dataset, which were manually edited for clarity. RESULTS Classification performance on the Inst. test achieved AUROC 0.82 using SVM and 0.81 using BlueBERT. Without cross-institution transfer learning, performance on the SF test was limited to an AUROC of 0.42 using SVM and 0.56 using BlueBERT. BlueBERT_TRANSFER, which was fine-tuned on both datasets, improved the performance on SF test to AUROC 0.78. Performance of SVM, and BlueBERT_TRANSFER models on the manually curated Inst. reports (AUROC 0.85 and 0.74) was similar to human performance (AUROC 0.81). CONCLUSION: In summary, we successfully developed cross-institution NLP models on incident report text from radiation oncology centers. These models were able to detect high-severity reports similarly to humans on a curated dataset.
Abstract (translated)
目的：事故报告是医疗保健中安全和质量改进的重要工具，但手动审查耗时且需要专业知识。在此，我们介绍了一种自然语言处理（NLP）筛选工具，用于检测两个机构中的放射肿瘤学高严重性事故报告。 方法与材料：我们使用了两个文本数据集来训练和评估我们的NLP模型：来自我们机构的7,094份报告以及IAEA SAFRON(SF) 的571份报告，所有这些报告都已由临床内容专家进行了严重程度评分。我们训练并评估了两种类型的模型：基线支持向量机（SVM）和BlueBERT，这是一种在PubMed摘要和住院患者数据上进行预训练的大型语言模型。为了测试我们的模型的泛化能力，我们在两个方面进行了评估。首先，在使用Inst.-train训练后对SF-test进行了评估。其次，我们训练了BlueBERT_TRANSFER 模型，该模型先是在 Inst.-train 上进行微调，然后在 SF-train上进行微调，最后在SF测试集上进行测试。为了进一步分析模型性能，我们也检查了我们机构数据集中手动编辑以提高清晰度的59份报告的一个子集。 结果：在Inst. 测试上的分类表现使用SVM达到AUROC 0.82，使用BlueBERT 达到 AUROC 0.81。没有跨机构迁移学习的情况下，在SF测试中的性能仅限于 SVM 的AUROC为 0.42 和 BlueBERT的AUROC为 0.56。经过两个数据集微调后的BlueBERT_TRANSFER 模型，将 SF 测试上的表现提高到 AUROC 0.78。在手动整理的Inst. 报告上，SVM和BlueBERT_TRANSFER模型（AUROC分别为0.85和0.74）的表现与人类表现相似（AUROC为0.81）。 结论：总的来说，我们成功地开发了跨机构NLP模型用于放射肿瘤学中心的事故报告文本。这些模型能够在整理后的数据集中类似地检测出高严重性报告。
URL
https://arxiv.org/abs/2509.13706
PDF
-
Latent Traits and Cross-Task Transfer: Deconstructing Dataset Interactions in LLM Fine-tuning
2025-09-17 01:45:42Shambhavi Krishna, Atharva Naik, Chaitali Agarwal, Sudharshan Govindan, Taesung Lee, Haw-Shiuan ChangAbstract
Large language models are increasingly deployed across diverse applications. This often includes tasks LLMs have not encountered during training. This implies that enumerating and obtaining the high-quality training data for all tasks is infeasible. Thus, we often need to rely on transfer learning using datasets with different characteristics, and anticipate out-of-distribution requests. Motivated by this practical need, we propose an analysis framework, building a transfer learning matrix and dimensionality reduction, to dissect these cross-task interactions. We train and analyze 10 models to identify latent abilities (e.g., Reasoning, Sentiment Classification, NLU, Arithmetic) and discover the side effects of the transfer learning. Our findings reveal that performance improvements often defy explanations based on surface-level dataset similarity or source data quality. Instead, hidden statistical factors of the source dataset, such as class distribution and generation length proclivities, alongside specific linguistic features, are actually more influential. This work offers insights into the complex dynamics of transfer learning, paving the way for more predictable and effective LLM adaptation.
Abstract (translated)
大型语言模型在各种应用中越来越广泛地部署。这通常包括那些在训练过程中未遇到的任务。这意味着为所有任务枚举并获取高质量的训练数据是不可行的。因此，我们常常需要依赖于使用具有不同特征的数据集进行迁移学习，并且要预料到分布外请求的情况。出于这一实际需求，我们提出了一种分析框架，构建一个转移学习矩阵和维度缩减，以剖析这些跨任务交互。通过训练并分析10个模型来识别潜在的能力（如推理、情感分类、自然语言理解、算术），我们发现迁移学习的副作用。我们的研究结果表明，性能提升往往无法用基于表面数据集相似性或源数据质量的解释来充分说明。相反，源数据集中隐藏的统计因素，例如类别分布和生成长度倾向，以及特定的语言特征实际上更具影响力。这项工作提供了对迁移学习复杂动态的理解，并为更可预测且有效的大型语言模型适应铺平了道路。
URL
https://arxiv.org/abs/2509.13624
PDF
-
Enhancing Speaker-Independent Dysarthric Speech Severity Classification with DSSCNet and Cross-Corpus Adaptation
2025-09-16 18:19:00Arnab Kumar Roy, Hemant Kumar Kathania, Paban SapkotaAbstract
Dysarthric speech severity classification is crucial for objective clinical assessment and progress monitoring in individuals with motor speech disorders. Although prior methods have addressed this task, achieving robust generalization in speaker-independent (SID) scenarios remains challenging. This work introduces DSSCNet, a novel deep neural architecture that combines Convolutional, Squeeze-Excitation (SE), and Residual network, helping it extract discriminative representations of dysarthric speech from mel spectrograms. The addition of SE block selectively focuses on the important features of the dysarthric speech, thereby minimizing loss and enhancing overall model performance. We also propose a cross-corpus fine-tuning framework for severity classification, adapted from detection-based transfer learning approaches. DSSCNet is evaluated on two benchmark dysarthric speech corpora: TORGO and UA-Speech under speaker-independent evaluation protocols: One-Speaker-Per-Severity (OSPS) and Leave-One-Speaker-Out (LOSO) protocols. DSSCNet achieves accuracies of 56.84% and 62.62% under OSPS and 63.47% and 64.18% under LOSO setting on TORGO and UA-Speech respectively outperforming existing state-of-the-art methods. Upon fine-tuning, the performance improves substantially, with DSSCNet achieving up to 75.80% accuracy on TORGO and 68.25% on UA-Speech in OSPS, and up to 77.76% and 79.44%, respectively, in LOSO. These results demonstrate the effectiveness and generalizability of DSSCNet for fine-grained severity classification across diverse dysarthric speech datasets.
Abstract (translated)
构音障碍语音严重程度分类对于患有运动性言语障碍的个体来说，是临床客观评估和进程监测的关键。尽管之前的方法已经解决了这一任务，但在说话人无关（SID）场景中实现稳健泛化仍然具有挑战性。本文介绍了一种新颖的深度神经网络架构DSSCNet，该架构结合了卷积、挤压-激励（SE）以及残差网络，有助于从梅尔频谱图中提取构音障碍语音的区分表示。通过添加SE模块，可以有针对性地关注构音障碍语音的重要特征，从而减少损失并提高整体模型性能。我们还提出了一种用于严重程度分类的跨语料库微调框架，该框架借鉴了基于检测的迁移学习方法。 DSSCNet在两个基准构音障碍语音数据集（TORGO和UA-Speech）上进行了评估，在说话人无关的评估协议下：单个说话人每一种严重程度（OSPS）和留一说话人法（LOSO）。在TORGO和UA-Speech数据集中，DSSCNet分别达到了56.84%和62.62%（OSPS）以及63.47%和64.18%（LOSO）的准确率，在这些任务上优于现有的最先进方法。经过微调后，性能显著提高：在TORGO数据集中DSSCNet分别达到了75.80%（OSPS）和77.76%（LOSO），而在UA-Speech数据集中则为68.25%（OSPS）和79.44%（LOSO）。这些结果证明了DSSCNet在跨多种构音障碍语音数据集进行精细程度分类时的有效性和泛化能力。
URL
https://arxiv.org/abs/2509.13442
PDF
-
Enhancing Smart Farming Through Federated Learning: A Secure, Scalable, and Efficient Approach for AI-Driven Agriculture
2025-09-15 18:57:32Ritesh Janga, Rushit DaveAbstract
The agricultural sector is undergoing a transformation with the integration of advanced technologies, particularly in data-driven decision-making. This work proposes a federated learning framework for smart farming, aiming to develop a scalable, efficient, and secure solution for crop disease detection tailored to the environmental and operational conditions of Minnesota farms. By maintaining sensitive farm data locally and enabling collaborative model updates, our proposed framework seeks to achieve high accuracy in crop disease classification without compromising data privacy. We outline a methodology involving data collection from Minnesota farms, application of local deep learning algorithms, transfer learning, and a central aggregation server for model refinement, aiming to achieve improved accuracy in disease detection, good generalization across agricultural scenarios, lower costs in communication and training time, and earlier identification and intervention against diseases in future implementations. We outline a methodology and anticipated outcomes, setting the stage for empirical validation in subsequent studies. This work comes in a context where more and more demand for data-driven interpretations in agriculture has to be weighed with concerns about privacy from farms that are hesitant to share their operational data. This will be important to provide a secure and efficient disease detection method that can finally revolutionize smart farming systems and solve local agricultural problems with data confidentiality. In doing so, this paper bridges the gap between advanced machine learning techniques and the practical, privacy-sensitive needs of farmers in Minnesota and beyond, leveraging the benefits of federated learning.
Abstract (translated)
农业部门正在通过集成先进科技，特别是在数据驱动决策方面进行转型。本文提出了一种用于智能农场的联邦学习框架，旨在为明尼苏达州农场量身定制一种可扩展、高效且安全的作物病害检测解决方案。通过在本地保留敏感的农场数据，并实现协作模型更新，我们提出的这一框架力求在不牺牲数据隐私的前提下，提高作物病害分类的准确性。 该方法涉及从明尼苏达州农场收集数据，应用本地深度学习算法，进行迁移学习以及利用中央聚合服务器来优化模型。其目标是提升疾病检测的准确率、增强跨农业场景的一般化能力、降低通信和训练时间的成本，并在未来的实施中实现早期病害识别与干预。 本文还概述了研究方法及预期成果，为后续实证验证奠定了基础。在此背景下，越来越多地需要对基于数据驱动的解释进行评估，同时考虑到那些不愿分享运营数据的农场对于隐私的担忧。这将有助于提供一种安全高效、能够最终革新智能农业系统的病害检测方式，并以数据保密性解决当地农业生产的问题。 通过这种方法，本文在高级机器学习技术与明尼苏达州乃至更广泛地区的农民的实际和隐私敏感需求之间建立了联系，充分利用了联邦学习的优势。
URL
https://arxiv.org/abs/2509.12363
PDF
-
The Quest for Universal Master Key Filters in DS-CNNs
2025-09-15 09:10:13Zahra Babaiee, Peyman M. Kiassari, Daniela Rus, Radu GrosuAbstract
A recent study has proposed the "Master Key Filters Hypothesis" for convolutional neural network filters. This paper extends this hypothesis by radically constraining its scope to a single set of just 8 universal filters that depthwise separable convolutional networks inherently converge to. While conventional DS-CNNs employ thousands of distinct trained filters, our analysis reveals these filters are predom- inantly linear shifts (ax+b) of our discovered universal set. Through systematic unsupervised search, we extracted these fundamental patterns across different architectures and datasets. Remarkably, networks initialized with these 8 unique frozen filters achieve over 80% ImageNet accuracy, and even outperform models with thousands of trainable parameters when applied to smaller datasets. The identified master key filters closely match Difference of Gaussians (DoGs), Gaussians, and their derivatives, structures that are not only fundamental to classical image processing but also strikingly similar to receptive fields in mammalian visual systems. Our findings provide compelling evidence that depthwise convolutional layers naturally gravitate toward this fundamental set of spatial operators regardless of task or architecture. This work offers new insights for understanding generalization and transfer learning through the universal language of these master key filters.
Abstract (translated)
最近的一项研究提出了卷积神经网络滤波器的“主密钥滤波假设”。本文在此基础上进行了扩展，通过激进地将该假设的范围限制为深度可分离卷积网络内在收敛到的一组仅有的8个通用滤波器来实现。尽管传统的DS-CNNs 使用了数千种不同的训练滤波器，但我们的分析表明，这些滤波器主要是我们发现的通用集合中线性变换（ax+b）的结果。通过系统的无监督搜索，我们在不同架构和数据集上提取了这些基本模式。值得注意的是，使用这8个独特的冻结滤波器初始化的网络在ImageNet数据集上的准确率超过了80%，甚至在应用于较小的数据集时也优于具有数千可训练参数的模型。 我们识别出的主密钥滤波器与高斯差（DoGs）、高斯函数及其导数非常接近，这些结构不仅是经典图像处理的基本组成部分，而且与哺乳动物视觉系统中的感受野惊人地相似。我们的发现提供了强有力的证据表明，深度卷积层无论在何种任务或架构下都会自然趋向于这一基本的空间操作集。 这项工作通过主密钥滤波器的通用语言为理解泛化和迁移学习提供了新的见解。
URL
https://arxiv.org/abs/2509.11711
PDF
-
Synthetic Dataset Evaluation Based on Generalized Cross Validation
2025-09-14 13:57:33Zhihang Song, Dingyi Yao, Ruibo Ming, Lihui Peng, Danya Yao, Yi ZhangAbstract
With the rapid advancement of synthetic dataset generation techniques, evaluating the quality of synthetic data has become a critical research focus. Robust evaluation not only drives innovations in data generation methods but also guides researchers in optimizing the utilization of these synthetic resources. However, current evaluation studies for synthetic datasets remain limited, lacking a universally accepted standard framework. To address this, this paper proposes a novel evaluation framework integrating generalized cross-validation experiments and domain transfer learning principles, enabling generalizable and comparable assessments of synthetic dataset quality. The framework involves training task-specific models (e.g., YOLOv5s) on both synthetic datasets and multiple real-world benchmarks (e.g., KITTI, BDD100K), forming a cross-performance matrix. Following normalization, a Generalized Cross-Validation (GCV) Matrix is constructed to quantify domain transferability. The framework introduces two key metrics. One measures the simulation quality by quantifying the similarity between synthetic data and real-world datasets, while another evaluates the transfer quality by assessing the diversity and coverage of synthetic data across various real-world scenarios. Experimental validation on Virtual KITTI demonstrates the effectiveness of our proposed framework and metrics in assessing synthetic data fidelity. This scalable and quantifiable evaluation solution overcomes traditional limitations, providing a principled approach to guide synthetic dataset optimization in artificial intelligence research.
Abstract (translated)
随着合成数据集生成技术的迅速发展，评估合成数据的质量已成为一个关键的研究焦点。稳健的评估不仅推动了数据生成方法的创新，还指导研究人员优化这些合成资源的应用。然而，目前对于合成数据集的评价研究仍然有限，并缺乏普遍接受的标准框架。为解决这一问题，本文提出了一种新的评估框架，该框架结合了广义交叉验证实验和领域转换学习原则，能够对合成数据的质量进行通用且可比较的评估。此框架涉及在合成数据集和多个真实世界基准（如KITTI、BDD100K）上训练特定任务的模型（例如YOLOv5s），从而形成一个跨性能矩阵。经过标准化处理后，构建广义交叉验证（GCV）矩阵来量化领域转换能力。该框架引入了两个关键指标：一个是通过量化合成数据与真实世界数据集之间的相似性来衡量模拟质量；另一个则是通过评估合成数据在各种现实场景中的多样性和覆盖范围来评定传输质量。基于虚拟KITTI的实验验证表明，我们提出的框架和度量方法能够有效地评估合成数据的真实度。这种可扩展且具量化性的评价解决方案克服了传统限制，为指导人工智能研究中合成数据集优化提供了一个原则性的方法。
URL
https://arxiv.org/abs/2509.11273
PDF
-
Revisiting Meter Tracking in Carnatic Music using Deep Learning Approaches
2025-09-14 12:33:34Satyajeet PrabhuAbstract
Beat and downbeat tracking, jointly referred to as Meter Tracking, is a fundamental task in Music Information Retrieval (MIR). Deep learning models have far surpassed traditional signal processing and classical machine learning approaches in this domain, particularly for Western (Eurogenetic) genres, where large annotated datasets are widely available. These systems, however, perform less reliably on underrepresented musical traditions. Carnatic music, a rich tradition from the Indian subcontinent, is renowned for its rhythmic intricacy and unique metrical structures (tÄlas). The most notable prior work on meter tracking in this context employed probabilistic Dynamic Bayesian Networks (DBNs). The performance of state-of-the-art (SOTA) deep learning models on Carnatic music, however, remains largely unexplored. In this study, we evaluate two models for meter tracking in Carnatic music: the Temporal Convolutional Network (TCN), a lightweight architecture that has been successfully adapted for Latin rhythms, and Beat This!, a transformer-based model designed for broad stylistic coverage without the need for post-processing. Replicating the experimental setup of the DBN baseline on the Carnatic Music Rhythm (CMR$_f$) dataset, we systematically assess the performance of these models in a directly comparable setting. We further investigate adaptation strategies, including fine-tuning the models on Carnatic data and the use of musically informed parameters. Results show that while off-the-shelf models do not always outperform the DBN, their performance improves substantially with transfer learning, matching or surpassing the baseline. These findings indicate that SOTA deep learning models can be effectively adapted to underrepresented traditions, paving the way for more inclusive and broadly applicable meter tracking systems.
Abstract (translated)
节拍和弱拍跟踪，统称为节奏跟踪（Meter Tracking），是音乐信息检索（MIR）中的基本任务之一。在西方（欧洲传统）音乐流派中，深度学习模型已经远远超越了传统的信号处理和经典机器学习方法，尤其是在有大量标注数据集可用的情况下。然而，这些系统在代表性不足的音乐传统上表现得不那么可靠。卡纳提克音乐是一种来自印度次大陆的丰富传统，以其复杂的节奏结构（tÄlas）而闻名。在此领域的先前研究中，最著名的工作之一是使用概率动态贝叶斯网络（DBNs）。然而，最先进的深度学习模型在卡纳提克音乐上的表现仍然未被充分探索。 在这项研究中，我们评估了两种用于卡纳提克音乐节奏跟踪的模型：时序卷积网络（TCN），这是一种轻量级架构，在拉丁节奏上已经成功地进行了适应；以及 Beat This!，这是一个基于变压器的模型，旨在覆盖广泛的风格，并且无需后期处理。我们通过在卡纳提克音乐节奏（CMR$_f$）数据集上复制DBN基线实验设置，系统评估了这些模型在一个可以直接比较的环境中表现。 此外，我们还探讨了适应策略，包括在卡纳提克数据上微调模型以及使用受音乐启发的信息参数。结果表明，虽然现成的模型并不总能超越DBN性能，但通过迁移学习其性能显著提升，达到了与基线相匹配或超过基线的水平。这些发现表明，最先进的深度学习模型可以有效地适应代表性不足的传统，为更包容和广泛适用的节奏跟踪系统铺平道路。
URL
https://arxiv.org/abs/2509.11241
PDF
-
Feature Space Topology Control via Hopkins Loss
2025-09-14 08:16:20Einari Vaaras, Manu AiraksinenAbstract
Feature space topology refers to the organization of samples within the feature space. Modifying this topology can be beneficial in machine learning applications, including dimensionality reduction, generative modeling, transfer learning, and robustness to adversarial attacks. This paper introduces a novel loss function, Hopkins loss, which leverages the Hopkins statistic to enforce a desired feature space topology, which is in contrast to existing topology-related methods that aim to preserve input feature topology. We evaluate the effectiveness of Hopkins loss on speech, text, and image data in two scenarios: classification and dimensionality reduction using nonlinear bottleneck autoencoders. Our experiments show that integrating Hopkins loss into classification or dimensionality reduction has only a small impact on classification performance while providing the benefit of modifying feature topology.
Abstract (translated)
特征空间拓扑指的是在特征空间中样本的组织方式。修改这种拓扑结构对于机器学习应用是有益的，包括维度缩减、生成模型、迁移学习以及对抗性攻击的鲁棒性等方面。本文介绍了一种新颖的损失函数——霍普金斯（Hopkins）损失，该函数利用霍普金斯统计量来强制执行所需的特征空间拓扑结构，这与现有的旨在保持输入特征拓扑的方法形成了对比。我们在分类和使用非线性瓶颈自编码器进行维度缩减这两种场景下，评估了霍普金斯损失在语音、文本及图像数据上的有效性。实验结果显示，在将霍普金斯损失集成到分类或维度缩减应用中时，对分类性能仅有很小的影响，但同时带来了修改特征拓扑的好处。
URL
https://arxiv.org/abs/2509.11154
PDF
-
Policy-Driven Transfer Learning in Resource-Limited Animal Monitoring
2025-09-13 22:26:51Nisha Pillai, Aditi Virupakshaiah, Harrison W. Smith, Amanda J. Ashworth, Prasanna Gowda, Phillip R. Owens, Adam R. Rivers, Bindu Nanduri, Mahalingam RamkumarAbstract
Animal health monitoring and population management are critical aspects of wildlife conservation and livestock management that increasingly rely on automated detection and tracking systems. While Unmanned Aerial Vehicle (UAV) based systems combined with computer vision offer promising solutions for non-invasive animal monitoring across challenging terrains, limited availability of labeled training data remains an obstacle in developing effective deep learning (DL) models for these applications. Transfer learning has emerged as a potential solution, allowing models trained on large datasets to be adapted for resource-limited scenarios such as those with limited data. However, the vast landscape of pre-trained neural network architectures makes it challenging to select optimal models, particularly for researchers new to the field. In this paper, we propose a reinforcement learning (RL)-based transfer learning framework that employs an upper confidence bound (UCB) algorithm to automatically select the most suitable pre-trained model for animal detection tasks. Our approach systematically evaluates and ranks candidate models based on their performance, streamlining the model selection process. Experimental results demonstrate that our framework achieves a higher detection rate while requiring significantly less computational time compared to traditional methods.
Abstract (translated)
动物健康监测和种群管理是野生动物保护和家畜管理中至关重要的方面，这些领域越来越依赖于自动化检测和跟踪系统。基于无人驾驶飞机（UAV）结合计算机视觉的系统为非侵入性动物监测提供了有前景的解决方案，尤其是在复杂地形条件下。然而，标记训练数据的有限可用性仍然是开发有效的深度学习（DL）模型以应对这些应用的主要障碍。迁移学习作为一种潜在的解决方案已崭露头角，它允许在大型数据集上进行训练的模型适应于资源受限的情景，例如数据量较小的情况。但是，大量的预训练神经网络架构使得选择最佳模型变得具有挑战性，特别是对于新进入该领域的研究人员而言。 本文提出了一种基于强化学习（RL）的迁移学习框架，该框架采用上置信界（UCB）算法自动选择最适合动物检测任务的预训练模型。我们的方法系统地评估和排名候选模型，并根据它们的表现对这些模型进行排序，从而简化了模型选择的过程。实验结果表明，与传统方法相比，我们提出的框架在提高检测率的同时显著减少了计算时间。
URL
https://arxiv.org/abs/2509.10995
PDF
-
Resource-Efficient Glioma Segmentation on Sub-Saharan MRI
2025-09-11 13:52:47Freedmore Sidume, Oumayma Soula, Joseph Muthui Wacira, YunFei Zhu, Abbas Rabiu Muhammad, Abderrazek Zeraii, Oluwaseun Kalejaye, Hajer Ibrahim, Olfa Gaddour, Brain Halubanza, Dong Zhang, Udunna C Anazodo, Confidence RaymondAbstract
Gliomas are the most prevalent type of primary brain tumors, and their accurate segmentation from MRI is critical for diagnosis, treatment planning, and longitudinal monitoring. However, the scarcity of high-quality annotated imaging data in Sub-Saharan Africa (SSA) poses a significant challenge for deploying advanced segmentation models in clinical workflows. This study introduces a robust and computationally efficient deep learning framework tailored for resource-constrained settings. We leveraged a 3D Attention UNet architecture augmented with residual blocks and enhanced through transfer learning from pre-trained weights on the BraTS 2021 dataset. Our model was evaluated on 95 MRI cases from the BraTS-Africa dataset, a benchmark for glioma segmentation in SSA MRI data. Despite the limited data quality and quantity, our approach achieved Dice scores of 0.76 for the Enhancing Tumor (ET), 0.80 for Necrotic and Non-Enhancing Tumor Core (NETC), and 0.85 for Surrounding Non-Functional Hemisphere (SNFH). These results demonstrate the generalizability of the proposed model and its potential to support clinical decision making in low-resource settings. The compact architecture, approximately 90 MB, and sub-minute per-volume inference time on consumer-grade hardware further underscore its practicality for deployment in SSA health systems. This work contributes toward closing the gap in equitable AI for global health by empowering underserved regions with high-performing and accessible medical imaging solutions.
Abstract (translated)
胶质瘤是最常见的原发性脑肿瘤类型，从MRI图像中对其进行准确分割对于诊断、治疗规划和纵向监测至关重要。然而，在撒哈拉以南非洲（SSA）地区高质量注释成像数据的稀缺给部署高级分割模型带来了重大挑战。本研究介绍了一种专为资源受限环境设计的强大且计算效率高的深度学习框架。我们采用了3D注意力UNet架构，并通过添加残差块进行了增强，同时利用在BraTS 2021数据集上预训练的权重进行迁移学习。我们的模型在包含95例MRI案例的BraTS-Africa数据集（用于SSA MRI数据中胶质瘤分割的标准基准）上进行了评估。尽管面临数据质量和数量的限制，我们的方法在增强肿瘤(ET)、坏死和非增强肿瘤核心(NETC)以及周围无功能半球(SNFH)上的Dice分数分别达到了0.76、0.80 和 0.85。这些结果展示了所提出模型的泛化能力及其支持低资源环境下临床决策制定的潜力。此外，该模型架构精简（约90MB），在消费级硬件上每体素推理时间不到一分钟，进一步突显了其在SSA卫生系统中的实用部署价值。 这项工作有助于弥合全球健康领域人工智能应用不平等的差距，通过为欠发达地区提供高性能且易于访问的医学影像解决方案。
URL
https://arxiv.org/abs/2509.09469
PDF
-
D-CAT: Decoupled Cross-Attention Transfer between Sensor Modalities for Unimodal Inference
2025-09-11 10:54:07Leen Daher, Zhaobo Wang, Malcolm MielleAbstract
Cross-modal transfer learning is used to improve multi-modal classification models (e.g., for human activity recognition in human-robot collaboration). However, existing methods require paired sensor data at both training and inference, limiting deployment in resource-constrained environments where full sensor suites are not economically and technically usable. To address this, we propose Decoupled Cross-Attention Transfer (D-CAT), a framework that aligns modality-specific representations without requiring joint sensor modality during inference. Our approach combines a self-attention module for feature extraction with a novel cross-attention alignment loss, which enforces the alignment of sensors' feature spaces without requiring the coupling of the classification pipelines of both modalities. We evaluate D-CAT on three multi-modal human activity datasets (IMU, video, and audio) under both in-distribution and out-of-distribution scenarios, comparing against uni-modal models. Results show that in in-distribution scenarios, transferring from high-performing modalities (e.g., video to IMU) yields up to 10% F1-score gains over uni-modal training. In out-of-distribution scenarios, even weaker source modalities (e.g., IMU to video) improve target performance, as long as the target model isn't overfitted on the training data. By enabling single-sensor inference with cross-modal knowledge, D-CAT reduces hardware redundancy for perception systems while maintaining accuracy, which is critical for cost-sensitive or adaptive deployments (e.g., assistive robots in homes with variable sensor availability). Code is available at this https URL.
Abstract (translated)
跨模态迁移学习被用于提升多模态分类模型（例如在人机协作中的人类活动识别）。然而，现有的方法需要在训练和推理阶段都有配对的传感器数据，这限制了其在无法经济和技术上支持全套传感器设备的资源受限环境中的部署。为解决这一问题，我们提出了解耦跨注意力迁移（Decoupled Cross-Attention Transfer, D-CAT），这是一种框架，在不需联合传感器模态的情况下，就能对特定于模式的表示进行对齐。我们的方法结合了自注意模块用于特征提取，并引入了一种新颖的跨注意对准损失，该损失强制要求在无需两种模式分类管道耦合的情况下，实现传感器特征空间的一致性。 我们在三个多模态人类活动数据集（IMU、视频和音频）上评估了D-CAT，在分布内和分布外场景下都进行了测试，并与单模态模型进行了比较。结果表明，在分布内场景中，从高表现力的模式转移（例如，从视频到IMU），可以比单模态训练高出高达10%的F1得分。在分布外场景中，即使是从较弱的源模式迁移过来（例如，从IMU到视频）也可以提高目标性能，只要目标模型没有过度拟合于训练数据即可。 通过使单一传感器推断具备跨模态知识，D-CAT减少了感知系统中的硬件冗余同时保持了准确性。这对于成本敏感或适应性部署至关重要（例如，在家中具有可变传感器可用性的辅助机器人）。代码可以在该链接处获取：[请在此插入实际的URL]
URL
https://arxiv.org/abs/2509.09747
PDF
