Transfer_Learning
-
Compressing Vision Transformers in Geospatial Transfer Learning with Manifold-Constrained Optimization
2026-01-12 18:41:39Thomas Snyder, H. Lexie Yang, Stefan Schnake, Steffen Schotth\"oferAbstract
Deploying geospatial foundation models on resource-constrained edge devices demands compact architectures that maintain high downstream performance. However, their large parameter counts and the accuracy loss often induced by compression limit practical adoption. In this work, we leverage manifold-constrained optimization framework DLRT to compress large vision transformer-based geospatial foundation models during transfer learning. By enforcing structured low-dimensional parameterizations aligned with downstream objectives, this approach achieves strong compression while preserving task-specific accuracy. We show that the method outperforms of-the-shelf low-rank methods as LoRA. Experiments on diverse geospatial benchmarks confirm substantial parameter reduction with minimal accuracy loss, enabling high-performing, on-device geospatial models.
Abstract (translated)
在资源受限的边缘设备上部署地理空间基础模型需要采用紧凑架构以保持高性能，但这些模型庞大的参数量以及压缩过程中常常导致的精度损失限制了其实用性。在这项工作中，我们利用流形约束优化框架DLRT，在迁移学习阶段对基于视觉变压器的大规模地理空间基础模型进行压缩。通过强制执行与下游目标相一致的结构化低维参数化，该方法在保持任务特定准确性的前提下实现了强大的压缩效果。实验结果表明，这种方法优于现成的低秩方法（如LoRA）。在多种地理空间基准上的实验证明了其能够在大幅减少参数量的同时，几乎不损失精度，从而能够实现高性能、设备内置的地理空间模型。
URL
https://arxiv.org/abs/2601.08882
PDF
-
AnyECG: Evolved ECG Foundation Model for Holistic Health Profiling
2026-01-12 16:03:52Jun Li, Hongling Zhu, Yujie Xiao, Qinghao Zhao, Yalei Ke, Gongzheng Tang, Guangkun Nie, Deyun Zhang, Jin Li, Canqing Yu, Shenda HongAbstract
Background: Artificial intelligence enabled electrocardiography (AI-ECG) has demonstrated the ability to detect diverse pathologies, but most existing models focus on single disease identification, neglecting comorbidities and future risk prediction. Although ECGFounder expanded cardiac disease coverage, a holistic health profiling model remains needed. Methods: We constructed a large multicenter dataset comprising 13.3 million ECGs from 2.98 million patients. Using transfer learning, ECGFounder was fine-tuned to develop AnyECG, a foundation model for holistic health profiling. Performance was evaluated using external validation cohorts and a 10-year longitudinal cohort for current diagnosis, future risk prediction, and comorbidity identification. Results: AnyECG demonstrated systemic predictive capability across 1172 conditions, achieving an AUROC greater than 0.7 for 306 diseases. The model revealed novel disease associations, robust comorbidity patterns, and future disease risks. Representative examples included high diagnostic performance for hyperparathyroidism (AUROC 0.941), type 2 diabetes (0.803), Crohn disease (0.817), lymphoid leukemia (0.856), and chronic obstructive pulmonary disease (0.773). Conclusion: The AnyECG foundation model provides substantial evidence that AI-ECG can serve as a systemic tool for concurrent disease detection and long-term risk prediction.
Abstract (translated)
背景：人工智能赋能的心电图（AI-ECG）展示了检测多种疾病的能力，但大多数现有的模型主要集中在单一疾病的识别上，忽略了共病和未来风险预测。尽管ECGFounder扩大了心脏疾病的覆盖范围，但仍需构建一个能够全面健康评估的基础模型。方法：我们构建了一个大型多中心数据集，包含298万名患者共计1330万份心电图记录。通过迁移学习，对ECGFounder进行微调以开发AnyECG，这是一种用于全面健康评估的基础模型。使用外部验证队列和一个长达十年的纵向队列来评估当前诊断、未来风险预测以及共病识别方面的性能。结果：AnyECG在1172种条件下的系统性预测能力表现出色，在306种疾病中达到了AUROC大于0.7的成绩。该模型揭示了新的疾病关联，稳健的共病模式和未来的疾病风险。代表性的例子包括高甲状旁腺功能亢进症（AUROC 0.941）、2型糖尿病（0.803）、克罗恩病（0.817）、淋巴性白血病（0.856）以及慢性阻塞性肺病（0.773）。结论：AnyECG基础模型提供了有力的证据，证明AI-ECG可以作为一个系统工具用于同时检测当前疾病和长期风险预测。
URL
https://arxiv.org/abs/2601.10748
PDF
-
Multi-modal MRI-Based Alzheimer's Disease Diagnosis with Transformer-based Image Synthesis and Transfer Learning
2026-01-09 22:26:38Jason QiuAbstract
Alzheimer's disease (AD) is a progressive neurodegenerative disorder in which pathological changes begin many years before the onset of clinical symptoms, making early detection essential for timely intervention. T1-weighted (T1w) Magnetic Resonance Imaging (MRI) is routinely used in clinical practice to identify macroscopic brain alterations, but these changes typically emerge relatively late in the disease course. Diffusion MRI (dMRI), in contrast, is sensitive to earlier microstructural abnormalities by probing water diffusion in brain tissue. dMRI metrics, including fractional anisotropy (FA) and mean diffusivity (MD), provide complementary information about white matter integrity and neurodegeneration. However, dMRI acquisitions are time-consuming and susceptible to motion artifacts, limiting their routine use in clinical populations. To bridge this gap, I propose a 3D TransUNet image synthesis framework that predicts FA and MD maps directly from T1w MRI. My model generates high-fidelity maps, achieving a structural similarity index (SSIM) exceeding 0.93 and a strong Pearson correlation (>0.94) with ground-truth dMRI. When integrated into a multi-modal diagnostic model, these synthetic features boost AD classification accuracy by 5% (78.75%->83.75%) and, most importantly, improve mild cognitive impairment (MCI) detection by 12.5%. This study demonstrates that high-quality diffusion microstructural information can be inferred from routinely acquired T1w MRI, effectively transferring the benefits of multi-modality imaging to settings where diffusion data are unavailable. By reducing scan time while preserving complementary structural and microstructural information, the proposed approach has the potential to improve the accessibility, efficiency, and accuracy of AD diagnosis in clinical practice.
Abstract (translated)
阿尔茨海默病（AD）是一种进行性神经退行性疾病，其病理变化早在临床症状出现前几年就开始了。因此，早期检测对于及时干预至关重要。T1加权（T1w）磁共振成像（MRI）在临床上被常规用于识别大脑的宏观改变，但这些变化通常是在疾病的晚期才显现出来。相比之下，弥散MRI（dMRI）通过探测脑组织中的水扩散来敏感地捕捉更早出现的微观结构异常。dMRI指标包括分数各向异性（FA）和平均扩散率（MD），提供了关于白质完整性和神经退行性变性的补充信息。然而，由于采集时间长且容易受到运动伪影的影响，弥散成像在临床人群中常规使用受到了限制。 为了弥补这一差距，我提出了一种基于3D TransUNet的图像合成框架，该框架能够直接从T1w MRI预测FA和MD图谱。我的模型生成了高保真的图谱，在结构相似性指数（SSIM）方面超过了0.93，并且与真实dMRI数据之间的皮尔逊相关系数大于0.94。当将这些合成特征整合到多模态诊断模型中时，它们能够提升AD分类的准确性5%（从78.75%提升至83.75%），更重要的是，在轻度认知障碍（MCI）检测方面提高了12.5%。 这项研究表明，可以通过常规获取的T1w MRI推断高质量的弥散微结构信息，从而在没有弥散数据的情况下也能实现多模态成像的优势。通过减少扫描时间同时保留互补的结构性和微观结构信息，所提出的这种方法有潜力提高阿尔茨海默病临床诊断中的可及性、效率和准确性。
URL
https://arxiv.org/abs/2601.11614
PDF
-
Multimodal In-context Learning for ASR of Low-resource Languages
2026-01-09 10:52:23Zhaolin Li, Jan NiehuesAbstract
Automatic speech recognition (ASR) still covers only a small fraction of the world's languages, mainly due to supervised data scarcity. In-context learning (ICL) with large language models (LLMs) addresses this problem, but prior work largely focuses on high-resource languages covered during training and text-only settings. This paper investigates whether speech LLMs can learn unseen languages with multimodal ICL (MICL), and how this learning can be used to improve ASR. We conduct experiments with two speech LLMs, Phi-4 and Qwen3-Omni, on three diverse endangered languages. Firstly, we find that MICL is effective for unseen languages, leveraging both speech and text modalities. We further show that cross-lingual transfer learning improves MICL efficiency on target languages without training on them. Moreover, we analyze attention patterns to interpret MICL mechanisms, and we observe layer-dependent preferences between audio and text context, with an overall bias towards text. Finally, we show that prompt-based ASR with speech LLMs performs poorly on unseen languages, motivating a simple ASR system that combines a stronger acoustic model with a speech LLM via MICL-based selection of acoustic hypotheses. Results show that MICL consistently improves ASR performance, and that cross-lingual transfer learning matches or outperforms corpus-trained language models without using target-language data. Our code is publicly available.
Abstract (translated)
这篇文档探讨了多模态上下文学习（MICL）在语音大语言模型（LLM）中的应用，特别是在处理低资源和未见过的语言方面的潜力。具体来说： 自动语音识别（ASR）目前仅覆盖世界一小部分语言，主要由于监督数据稀缺的问题。而在大型语言模型中进行的上下文学习（ICL）可以解决这个问题，但现有研究多集中在训练期间使用的高资源语言以及单一文本环境上。本文探讨了是否可以通过使用基于语音的大语言模型和多模态上下文学习（MICL）来学习未见过的语言，并研究这种学习如何能提高ASR性能。 研究人员对两个不同的语音大语言模型——Phi-4和Qwen3-Omni进行了实验，涉及三种濒危且多样化的语言。主要发现包括： 1. **多模态上下文学习的有效性**：结果显示，通过利用声音和文本的双模式信息，MICL在处理未见过的语言时是有效的。 2. **跨语种迁移学习的优势**：研究还表明，在目标语言上进行训练之前，可以通过跨语种转移学习提高MICL效率。这意味着可以在没有特定语言数据的情况下改进该语言上的ASR性能。 3. **注意力模式的分析**：研究人员还通过分析模型中不同层对音频和文本上下文的偏好来解释多模态上下文学习机制的工作原理，并观察到总体而言，模型更倾向于依赖文本信息而非语音信息。 4. **基于提示的ASR与MICL结合系统的性能对比**：研究发现，在处理未见过的语言时，仅依靠提示进行自动语音识别（即直接利用语言模型生成转录）的表现较差。因此，作者提出了一种简单的系统设计，该系统结合了一个更强的声学模型和一个通过MICL基于提示选择最佳假设的大语言模型。 实验结果表明，无论是在多模态上下文学习本身还是在通过跨语种迁移学习改进目标语言上的ASR性能方面，方法都能持续提高自动语音识别的效果。特别是在没有针对特定语言训练数据的情况下，跨语种转移学习的表现可以与使用该语言的语料库进行训练的语言模型相媲美或超越。 研究代码对公众开放，这有助于其他研究人员在低资源和濒危语言领域中进一步探索多模态上下文学习的应用潜力。
URL
https://arxiv.org/abs/2601.05707
PDF
-
An Empirical Study on Knowledge Transfer under Domain and Label Shifts in 3D LiDAR Point Clouds
2026-01-09 06:00:05Subeen Lee, Siyeong Lee, Namil Kim, Jaesik ChoiAbstract
For 3D perception systems to be practical in real-world applications -- from autonomous driving to embodied AI -- models must adapt to continuously evolving object definitions and sensor domains. Yet, research on continual and transfer learning in 3D point cloud perception remains underexplored compared to 2D vision -- particularly under simultaneous domain and label shifts. To address this gap, we propose the RObust Autonomous driving under Dataset shifts (ROAD) benchmark, a comprehensive evaluation suite for LiDAR-based object classification that explicitly accounts for domain shifts as well as three key forms of label evolution: class split, class expansion, and class insertion. Using large-scale datasets (Waymo, NuScenes, Argoverse2), we evaluate zero-shot transfer, linear probe, and CL, and analyze the impact of backbone architectures, training objectives, and CL methods. Our findings reveal limitations of existing approaches under realistic shifts and establish strong baselines for future research in robust 3D perception.
Abstract (translated)
为了使三维感知系统在现实世界应用（从自动驾驶到具身人工智能）中变得实用，模型必须适应不断变化的对象定义和传感器领域。然而，在三维点云感知中的连续学习和迁移学习的研究相比二维视觉仍然相对较少，特别是在同时出现领域和标签转换的情况下。为了解决这一不足，我们提出了“ROAD”基准测试（Robust Autonomous Driving Under Dataset Shifts的简称），这是一个全面评估基于激光雷达的对象分类性能的评测套件，该套件明确考虑了领域转换以及三种关键形式的标签演变：类别分裂、类别扩展和类别插入。利用大规模数据集（Waymo、NuScenes 和 Argoverse2），我们评估了零样本迁移学习、线性探测和持续学习方法，并分析了骨干架构、训练目标和持续学习方法的影响。我们的研究结果揭示了现有方法在现实世界转换中的局限性，并为未来稳健的三维感知研究建立了强有力的基础。
URL
https://arxiv.org/abs/2601.07855
PDF
-
Training a Custom CNN on Five Heterogeneous Image Datasets
2026-01-08 08:44:17Anika Tabassum, Tasnuva Mahazabin Tuba, Nafisa NazninAbstract
Deep learning has transformed visual data analysis, with Convolutional Neural Networks (CNNs) becoming highly effective in learning meaningful feature representations directly from images. Unlike traditional manual feature engineering methods, CNNs automatically extract hierarchical visual patterns, enabling strong performance across diverse real-world contexts. This study investigates the effectiveness of CNN-based architectures across five heterogeneous datasets spanning agricultural and urban domains: mango variety classification, paddy variety identification, road surface condition assessment, auto-rickshaw detection, and footpath encroachment monitoring. These datasets introduce varying challenges, including differences in illumination, resolution, environmental complexity, and class imbalance, necessitating adaptable and robust learning models. We evaluate a lightweight, task-specific custom CNN alongside established deep architectures, including ResNet-18 and VGG-16, trained both from scratch and using transfer learning. Through systematic preprocessing, augmentation, and controlled experimentation, we analyze how architectural complexity, model depth, and pre-training influence convergence, generalization, and performance across datasets of differing scale and difficulty. The key contributions of this work are: (1) the development of an efficient custom CNN that achieves competitive performance across multiple application domains, and (2) a comprehensive comparative analysis highlighting when transfer learning and deep architectures provide substantial advantages, particularly in data-constrained environments. These findings offer practical insights for deploying deep learning models in resource-limited yet high-impact real-world visual classification tasks.
Abstract (translated)
深度学习已经彻底改变了视觉数据分析，卷积神经网络（CNN）尤其擅长直接从图像中学习有意义的特征表示。与传统的手动特征工程方法不同，CNN能够自动提取分层的视觉模式，在各种现实世界的背景下表现出强大的性能。本研究探讨了基于CNN架构在五个异构数据集上的有效性，这些数据集涵盖了农业和城市领域：芒果品种分类、水稻品种识别、路面状况评估、机动三轮车检测以及人行道侵占监控。这些数据集引入了不同的挑战，包括光照条件的变化、分辨率差异、环境复杂性和类别不平衡等问题，这需要灵活且鲁棒的学习模型。我们评估了一个轻量级的任务特定自定义CNN，并与现有的深度架构进行比较，包括从头开始训练和使用迁移学习的ResNet-18和VGG-16。通过系统的预处理、增强和控制实验，我们分析了架构复杂性、模型深度以及预训练如何影响不同规模和难度的数据集的收敛性、泛化性和性能。 这项工作的主要贡献包括：（1）开发了一种高效的定制CNN，在多个应用领域中实现了竞争性的表现；（2）通过全面比较分析展示了在数据有限环境下，迁移学习和深层架构提供了显著的优势。这些发现为在资源受限但具有重要影响的现实世界视觉分类任务中部署深度学习模型提供了解决方案的实际见解。
URL
https://arxiv.org/abs/2601.04727
PDF
-
Performance Analysis of Image Classification on Bangladeshi Datasets
2026-01-07 21:15:16Mohammed Sami Khan, Fabiha Muniat, Rowzatul ZannatAbstract
Convolutional Neural Networks (CNNs) have demonstrated remarkable success in image classification tasks; however, the choice between designing a custom CNN from scratch and employing established pre-trained architectures remains an important practical consideration. In this work, we present a comparative analysis of a custom-designed CNN and several widely used deep learning architectures, including VGG-16, ResNet-50, and MobileNet, for an image classification task. The custom CNN is developed and trained from scratch, while the popular architectures are employed using transfer learning under identical experimental settings. All models are evaluated using standard performance metrics such as accuracy, precision, recall, and F1-score. Experimental results show that pre-trained CNN architectures consistently outperform the custom CNN in terms of classification accuracy and convergence speed, particularly when training data is limited. However, the custom CNN demonstrates competitive performance with significantly fewer parameters and reduced computational complexity. This study highlights the trade-offs between model complexity, performance, and computational efficiency, and provides practical insights into selecting appropriate CNN architectures for image classification problems.
Abstract (translated)
卷积神经网络（CNN）在图像分类任务中表现出卓越的成功；然而，是设计自定义的CNN还是使用已有的预训练架构仍然是一个重要的实际考虑因素。在这项工作中，我们对一个定制设计的CNN和几个广泛使用的深度学习架构（包括VGG-16、ResNet-50和MobileNet）进行了比较分析，以进行图像分类任务。该定制的CNN从头开始开发并训练，而流行的架构则在相同的实验设置下通过迁移学习使用。所有模型均根据标准性能指标（如准确率、精确率、召回率和F1分数）进行评估。实验结果表明，在数据集有限的情况下，预训练的CNN架构在分类精度和收敛速度方面始终优于自定义的CNN。然而，定制的CNN在参数数量较少且计算复杂度较低时表现出相当的竞争性能。这项研究强调了模型复杂性、性能和计算效率之间的权衡，并为选择适合图像分类问题的适当CNN架构提供了实用见解。
URL
https://arxiv.org/abs/2601.04397
PDF
-
Comparative Analysis of Custom CNN Architectures versus Pre-trained Models and Transfer Learning: A Study on Five Bangladesh Datasets
2026-01-07 19:36:41Ibrahim Tanvir (University of Dhaka), Alif Ruslan (University of Dhaka), Sartaj Solaiman (University of Dhaka)Abstract
This study presents a comprehensive comparative analysis of custom-built Convolutional Neural Networks (CNNs) against popular pre-trained architectures (ResNet-18 and VGG-16) using both feature extraction and transfer learning approaches. We evaluated these models across five diverse image classification datasets from Bangladesh: Footpath Vision, Auto Rickshaw Detection, Mango Image Classification, Paddy Variety Recognition, and Road Damage Detection. Our experimental results demonstrate that transfer learning with fine-tuning consistently outperforms both custom CNNs built from scratch and feature extraction methods, achieving accuracy improvements ranging from 3% to 76% across different datasets. Notably, ResNet-18 with fine-tuning achieved perfect 100% accuracy on the Road Damage BD dataset. While custom CNNs offer advantages in model size (3.4M parameters vs. 11-134M for pre-trained models) and training efficiency on simpler tasks, pre-trained models with transfer learning provide superior performance, particularly on complex classification tasks with limited training data. This research provides practical insights for practitioners in selecting appropriate deep learning approaches based on dataset characteristics, computational resources, and performance requirements.
Abstract (translated)
这项研究提供了一个全面的比较分析，对比了自定义构建的卷积神经网络（CNN）与流行的预训练架构（ResNet-18和VGG-16），使用特征提取和迁移学习两种方法。我们对五种来自孟加拉国的不同图像分类数据集进行了这些模型的评估：Footpath Vision、Auto Rickshaw Detection、Mango Image Classification、Paddy Variety Recognition以及Road Damage Detection。 我们的实验结果表明，通过微调进行迁移学习在所有情况下都优于从头开始构建的自定义CNN和特征提取方法，在不同的数据集中准确率提高了3%到76%。值得注意的是，经过微调后的ResNet-18模型在Road Damage BD数据集上实现了完美的100%精度。 尽管从头开始构建的CNN在网络大小（参数量为3.4M）和处理较为简单任务时的训练效率方面具有优势，但预训练模型结合迁移学习方法能够提供更优的表现，尤其是在处理复杂分类任务且训练数据有限的情况下。这项研究为实践者提供了基于数据集特性、计算资源以及性能需求来选择适当的深度学习策略的实际见解。
URL
https://arxiv.org/abs/2601.04352
PDF
-
Stuttering-Aware Automatic Speech Recognition for Indonesian Language
2026-01-07 09:21:12Fadhil Muhammad, Alwin Djuliansah, Adrian Aryaputra Hamzah, Kurniawati AzizahAbstract
Automatic speech recognition systems have achieved remarkable performance on fluent speech but continue to degrade significantly when processing stuttered speech, a limitation that is particularly acute for low-resource languages like Indonesian where specialized datasets are virtually non-existent. To overcome this scarcity, we propose a data augmentation framework that generates synthetic stuttered audio by injecting repetitions and prolongations into fluent text through a combination of rule-based transformations and large language models followed by text-to-speech synthesis. We apply this synthetic data to fine-tune a pre-trained Indonesian Whisper model using transfer learning, enabling the architecture to adapt to dysfluent acoustic patterns without requiring large-scale real-world recordings. Our experiments demonstrate that this targeted synthetic exposure consistently reduces recognition errors on stuttered speech while maintaining performance on fluent segments, validating the utility of synthetic data pipelines for developing more inclusive speech technologies in under-represented languages.
Abstract (translated)
自动语音识别系统在处理流畅的语音时已经取得了显著的成绩，但在处理口吃语音时性能仍然大幅下降，这一局限性在印尼语等资源匮乏的语言中尤为严重，因为这些语言缺乏专门的数据集。为了克服这种稀缺问题，我们提出了一种数据增强框架，该框架通过结合规则基础变换和大型语言模型来生成合成的口吃音频，在流畅文本中插入重复和延长部分，随后再进行语音合成。我们将这一合成数据应用于使用迁移学习对预训练的印尼语Whisper模型进行微调，使架构能够在不依赖大规模真实世界录音的情况下适应口吃的声学模式。实验表明，这种针对性的合成暴露方法在减少口吃语音识别错误方面具有持续性效果，并且保持了对流畅部分的表现力，从而验证了合成数据管道在开发更包容的语言技术方面的实用性，尤其是在代表性不足的语言中。
URL
https://arxiv.org/abs/2601.03727
PDF
-
A Comparative Study of Custom CNNs, Pre-trained Models, and Transfer Learning Across Multiple Visual Datasets
2026-01-05 16:26:32Annoor Sharara AkhandAbstract
Convolutional Neural Networks (CNNs) are a standard approach for visual recognition due to their capacity to learn hierarchical representations from raw pixels. In practice, practitioners often choose among (i) training a compact custom CNN from scratch, (ii) using a large pre-trained CNN as a fixed feature extractor, and (iii) performing transfer learning via partial or full fine-tuning of a pre-trained backbone. This report presents a controlled comparison of these three paradigms across five real-world image classification datasets spanning road-surface defect recognition, agricultural variety identification, fruit/leaf disease recognition, pedestrian walkway encroachment recognition, and unauthorized vehicle recognition. Models are evaluated using accuracy and macro F1-score, complemented by efficiency metrics including training time per epoch and parameter counts. The results show that transfer learning consistently yields the strongest predictive performance, while the custom CNN provides an attractive efficiency--accuracy trade-off, especially when compute and memory budgets are constrained.
Abstract (translated)
卷积神经网络（CNN）由于其从原始像素中学习分层表示的能力，成为视觉识别的标准方法。在实践中，从业人员通常会在这三种方式之间进行选择：(i) 从头开始训练紧凑型自定义的 CNN；(ii) 使用大型预训练的 CNN 作为固定的特征提取器；以及 (iii) 通过部分或完全微调预训练的基础模型来进行迁移学习。本报告对这三种范式进行了严格的比较，涵盖五个真实世界图像分类数据集，这些数据集涉及道路表面缺陷识别、农业品种识别、水果/叶片疾病识别、人行道侵入识别和未经授权的车辆识别。评估使用了准确率和宏F1分数指标，并辅以效率度量，包括每个训练周期的时间以及参数数量。结果表明，在预测性能方面，迁移学习始终表现出色；而定制 CNN 在计算资源（如计算能力和内存）受限的情况下提供了高效性和准确性之间的良好权衡。
URL
https://arxiv.org/abs/2601.02246
PDF
-
MambaFormer: Token-Level Guided Routing Mixture-of-Experts for Accurate and Efficient Clinical Assistance
2026-01-03 19:01:33Hamad Khan (Artificial Intelligence Lab, Department of Computer Systems Engineering, University of Engineering,Applied Sciences), Saddam Hussain Khan (Artificial Intelligence Lab, Department of Computer Systems Engineering, University of Engineering,Applied Sciences)Abstract
The deployment of large language models (LLMs) in real-world clinical applications is constrained by the fundamental trade-off between computational cost and the efficiency of linear-time models. To address this, we propose an LLM-based MambaFormer hybrid Mixture-of-Experts (MoE) framework for efficient medical question-answering (QA) and clinical assistance. The MambaFormer employs a lightweight gating mechanism that performs token-level dynamic routing to a customized Transformer expert (ET5) for short, complex queries or to a State Space Model expert (EMamba) for long, high-throughput sequences. The customized EMamba and ET5 models are tailored to accommodate input sequence dimensionality, embedding structure, sequence length, and target-specific output heads, and are fine-tuned through transfer learning on a new, custom-designed DentalQA dataset. Moreover, intelligent routing decisions are driven by the contextual complexity of token embeddings, normalized sequence length, and domain-aware features, thereby enforcing a Pareto-optimal trade-off between inference latency and prediction accuracy. Furthermore, a novel utility-guided multi-objective loss jointly optimizes decisions, router parameters, routing behavior, expert utilization, and computational cost by adaptively regulating token-level expert activation. Finally, the proposed MambaFormer is cross-validated (holdout) for medical QA on the new, custom-designed DentalQA and PubMedQA datasets and compared with state-of-the-art techniques. The proposed MambaFormer outperforms (BERTScore = 0.9180) with ultra-low latency (0.077 s), delivering a 24.4 speedup over T5-Large and establishing a scalable solution for resource-constrained clinical deployment.
Abstract (translated)
在现实世界的临床应用中部署大型语言模型（LLMs）受到计算成本和线性时间模型效率之间基本权衡的限制。为了解决这个问题，我们提出了一种基于LLM的MambaFormer混合专家组合(Mixture-of-Experts, MoE)框架，用于高效的医疗问答(QA)和临床辅助。MambaFormer采用轻量级门控机制，在处理简短复杂查询时将令牌动态路由到定制化的Transformer专家（ET5），在处理长序列或高吞吐量序列时则将其路由至状态空间模型专家（EMamba）。 定制的EMamba和ET5模型经过调整，以适应输入序列维度、嵌入结构、序列长度以及特定目标的输出头，并通过迁移学习对新的自设计牙科QA数据集进行微调。此外，智能路由决策由令牌嵌入的上下文复杂性、归一化序列长度及领域感知特征驱动，在推理延迟和预测准确性之间实现了帕累托最优权衡。同时，一种新颖的目标导向多目标损失函数通过调节令牌级专家激活来共同优化决策、路由器参数、路由行为、专家利用度以及计算成本。 最终，所提出的MambaFormer在新的自设计牙科QA数据集和PubMedQA数据集中进行交叉验证，并与最先进的技术进行了比较。该方法（BERTScore = 0.9180）在极低延迟(0.077秒)下表现出色，比T5-Large快24.4倍，在资源受限的临床部署中提供了一种可扩展的解决方案。
URL
https://arxiv.org/abs/2601.01260
PDF
-
Evolving CNN Architectures: From Custom Designs to Deep Residual Models for Diverse Image Classification and Detection Tasks
2026-01-03 07:45:08Mahmudul Hasan, Mabsur Fatin Bin HossainAbstract
This paper presents a comparative study of a custom convolutional neural network (CNN) architecture against widely used pretrained and transfer learning CNN models across five real-world image datasets. The datasets span binary classification, fine-grained multiclass recognition, and object detection scenarios. We analyze how architectural factors, such as network depth, residual connections, and feature extraction strategies, influence classification and localization performance. The results show that deeper CNN architectures provide substantial performance gains on fine-grained multiclass datasets, while lightweight pretrained and transfer learning models remain highly effective for simpler binary classification tasks. Additionally, we extend the proposed architecture to an object detection setting, demonstrating its adaptability in identifying unauthorized auto-rickshaws in real-world traffic scenes. Building upon a systematic analysis of custom CNN architectures alongside pretrained and transfer learning models, this study provides practical guidance for selecting suitable network designs based on task complexity and resource constraints.
Abstract (translated)
本文介绍了一种定制卷积神经网络（CNN）架构与五个真实世界图像数据集上广泛使用的预训练和迁移学习CNN模型之间的比较研究。这些数据集涵盖了二元分类、细粒度多类识别以及物体检测等场景。我们分析了诸如网络深度、残差连接和特征提取策略等结构因素如何影响分类及定位性能。结果显示，更深的CNN架构在细粒度多类数据集中提供了显著的性能提升，而轻量级的预训练和迁移学习模型对于更简单的二元分类任务仍然非常有效。此外，我们将提出的架构扩展到物体检测场景中，展示了其能够适应识别真实交通场景中的未经授权使用的小型三轮摩托车（auto-rickshaws）。通过对定制CNN架构与预训练及迁移学习模型进行系统的分析，本研究为基于任务复杂度和资源限制选择合适的网络设计提供了实用指导。
URL
https://arxiv.org/abs/2601.01099
PDF
-
Evaluating transfer learning strategies for improving dairy cattle body weight prediction in small farms using depth-image and point-cloud data
2026-01-03 02:41:54Jin Wang, Angelo De Castro, Yuxi Zhang, Lucas Basolli Borsatto, Yuechen Guo, Victoria Bastos Primo, Ana Beatriz Montevecchio Bernardino, Gota Morota, Ricardo C Chebel, Haipeng YuAbstract
Computer vision provides automated, non-invasive, and scalable tools for monitoring dairy cattle, thereby supporting management, health assessment, and phenotypic data collection. Although transfer learning is commonly used for predicting body weight from images, its effectiveness and optimal fine-tuning strategies remain poorly understood in livestock applications, particularly beyond the use of pretrained ImageNet or COCO weights. In addition, while both depth images and three-dimensional point-cloud data have been explored for body weight prediction, direct comparisons of these two modalities in dairy cattle are limited. Therefore, the objectives of this study were to 1) evaluate whether transfer learning from a large farm enhances body weight prediction on a small farm with limited data, and 2) compare the predictive performance of depth-image- and point-cloud-based approaches under three experimental designs. Top-view depth images and point-cloud data were collected from 1,201, 215, and 58 cows at large, medium, and small dairy farms, respectively. Four deep learning models were evaluated: ConvNeXt and MobileViT for depth images, and PointNet and DGCNN for point clouds. Transfer learning markedly improved body weight prediction on the small farm across all four models, outperforming single-source learning and achieving gains comparable to or greater than joint learning. These results indicate that pretrained representations generalize well across farms with differing imaging conditions and dairy cattle populations. No consistent performance difference was observed between depth-image- and point-cloud-based models. Overall, these findings suggest that transfer learning is well suited for small farm prediction scenarios where cross-farm data sharing is limited by privacy, logistical, or policy constraints, as it requires access only to pretrained model weights rather than raw data.
Abstract (translated)
计算机视觉提供了自动化、无创性和可扩展的工具，用于监测乳牛，从而支持管理、健康评估和表型数据收集。尽管在从图像预测体重时通常使用迁移学习，但其在畜牧业应用中的有效性和最佳微调策略仍然在很大程度上不为人知，尤其是在预训练ImageNet或COCO权重之外的应用中。此外，虽然深度图像和三维点云数据已被探索用于体重预测，但在乳牛上的这两种模式之间的直接比较是有限的。因此，本研究的目标是：1）评估来自大型农场的迁移学习是否可以增强在小型农场上进行体重预测的能力，特别是在数据较少的情况下；2）在三种实验设计下，对比基于深度图像和点云的方法的预测性能。 研究人员从大型、中型和小型乳牛场分别收集了1,201头、215头和58头奶牛的俯视图深度图像和点云数据。评估了四种深度学习模型：用于深度图像的ConvNeXt和MobileViT，以及用于点云的PointNet和DGCNN。迁移学习显著提升了小型农场上所有四个模型的体重预测能力，优于单源学习，并且在联合学习中的增益相当或更大。这些结果表明，预训练表示能够很好地跨不同成像条件和乳牛群体的农场进行泛化。未观察到深度图像和点云方法之间的性能差异。 总体而言，这些发现表明迁移学习非常适合于小型农场预测场景，在这种情况下，由于隐私、物流或政策限制而无法共享跨农场数据的情况中，因为其只需要访问预训练模型权重而不是原始数据。
URL
https://arxiv.org/abs/2601.01044
PDF
-
Quality Detection of Stored Potatoes via Transfer Learning: A CNN and Vision Transformer Approach
2026-01-02 11:10:55Shrikant Kapse, Priyankkumar Dhrangdhariya, Priya Kedia, Manasi Patwardhan, Shankar Kausley, Soumyadipta Maiti, Beena Rai, Shirish KarandeAbstract
Image-based deep learning provides a non-invasive, scalable solution for monitoring potato quality during storage, addressing key challenges such as sprout detection, weight loss estimation, and shelf-life prediction. In this study, images and corresponding weight data were collected over a 200-day period under controlled temperature and humidity conditions. Leveraging powerful pre-trained architectures of ResNet, VGG, DenseNet, and Vision Transformer (ViT), we designed two specialized models: (1) a high-precision binary classifier for sprout detection, and (2) an advanced multi-class predictor to estimate weight loss and forecast remaining shelf-life with remarkable accuracy. DenseNet achieved exceptional performance, with 98.03% accuracy in sprout detection. Shelf-life prediction models performed best with coarse class divisions (2-5 classes), achieving over 89.83% accuracy, while accuracy declined for finer divisions (6-8 classes) due to subtle visual differences and limited data per class. These findings demonstrate the feasibility of integrating image-based models into automated sorting and inventory systems, enabling early identification of sprouted potatoes and dynamic categorization based on storage stage. Practical implications include improved inventory management, differential pricing strategies, and reduced food waste across supply chains. While predicting exact shelf-life intervals remains challenging, focusing on broader class divisions ensures robust performance. Future research should aim to develop generalized models trained on diverse potato varieties and storage conditions to enhance adaptability and scalability. Overall, this approach offers a cost-effective, non-destructive method for quality assessment, supporting efficiency and sustainability in potato storage and distribution.
Abstract (translated)
基于图像的深度学习为监测储存期间土豆质量提供了一种非侵入性且可扩展的解决方案，解决了诸如发芽检测、重量损失估计和保质期预测等关键挑战。在这项研究中，在受控温度和湿度条件下收集了持续200天的图像和相应的重量数据。我们利用ResNet、VGG、DenseNet和Vision Transformer（ViT）这些强大的预训练架构，设计了两个专门化的模型：一是用于发芽检测的高精度二元分类器；二是用于估计重量损失并预测剩余保质期的先进多类预测器，并且它们都表现出非常高的准确性。 实验结果表明，DenseNet在发芽检测方面取得了优异的表现，准确率达到98.03%。对于保质期预测模型，在粗略类别划分（2-5个类别）时表现最佳，准确率超过89.83%，但在更细的类别划分（6-8个类别）中，由于细微的视觉差异和每个类别的数据有限，其准确性有所下降。 这些发现证明了将基于图像的模型整合到自动化分拣和库存系统中的可行性，能够实现对发芽土豆的早期识别，并根据储存阶段进行动态分类。实际应用包括改进库存管理、差异化定价策略以及减少供应链中的食品浪费。虽然预测精确的保质期间隔仍然具有挑战性，但专注于更广泛的类别划分可以确保模型性能稳定。 未来的研究应该致力于开发适用于各种土豆品种和存储条件的一般化模型，以增强适应性和可扩展性。总体而言，这种方法提供了一种低成本且非破坏性的质量评估方法，支持了土豆储存与分销的效率和可持续性。
URL
https://arxiv.org/abs/2601.00645
PDF
-
Noise-Aware Named Entity Recognition for Historical VET Documents
2026-01-01 21:43:35Alexander M. Esser, Jens D\"orpinghausAbstract
This paper addresses Named Entity Recognition (NER) in the domain of Vocational Education and Training (VET), focusing on historical, digitized documents that suffer from OCR-induced noise. We propose a robust NER approach leveraging Noise-Aware Training (NAT) with synthetically injected OCR errors, transfer learning, and multi-stage fine-tuning. Three complementary strategies, training on noisy, clean, and artificial data, are systematically compared. Our method is one of the first to recognize multiple entity types in VET documents. It is applied to German documents but transferable to arbitrary languages. Experimental results demonstrate that domain-specific and noise-aware fine-tuning substantially increases robustness and accuracy under noisy conditions. We provide publicly available code for reproducible noise-aware NER in domain-specific contexts.
Abstract (translated)
本文研究了职业教育与培训（VET）领域中的命名实体识别（NER），重点关注因光学字符识别（OCR）导致噪声的历史、数字化文档。我们提出了一种稳健的NER方法，利用带有合成注入OCR错误的Noise-Aware Training (NAT)、迁移学习和多阶段微调技术。文中系统地比较了三种互补策略：在有噪音数据、干净数据以及人工生成的数据上进行训练的效果。我们的方法是首次能够在VET文档中识别多种实体类型的模型，适用于德国文件但也可应用于任意语言的文本。实验结果表明，在噪声条件下，领域特定且具有噪声意识的微调显著提高了鲁棒性和准确性。我们提供了公开可用的代码以实现可重复的领域特异性噪声感知NER研究。
URL
https://arxiv.org/abs/2601.00488
PDF
-
The Impact of Lesion Focus on the Performance of AI-Based Melanoma Classification
2026-01-01 14:17:32Tanay DondeAbstract
Melanoma is the most lethal subtype of skin cancer, and early and accurate detection of this disease can greatly improve patients' outcomes. Although machine learning models, especially convolutional neural networks (CNNs), have shown great potential in automating melanoma classification, their diagnostic reliability still suffers due to inconsistent focus on lesion areas. In this study, we analyze the relationship between lesion attention and diagnostic performance, involving masked images, bounding box detection, and transfer learning. We used multiple explainability and sensitivity analysis approaches to investigate how well models aligned their attention with lesion areas and how this alignment correlated with precision, recall, and F1-score. Results showed that models with a higher focus on lesion areas achieved better diagnostic performance, suggesting the potential of interpretable AI in medical diagnostics. This study provides a foundation for developing more accurate and trustworthy melanoma classification models in the future.
Abstract (translated)
黑色素瘤是皮肤癌中最致命的一种，该疾病的早期和准确检测能够极大地改善患者的预后。尽管机器学习模型（尤其是卷积神经网络CNN）在自动化黑色素瘤分类方面展现出巨大潜力，但由于对病变区域的关注度不一致，其诊断可靠性仍然存在不足。在这项研究中，我们分析了病变关注与诊断性能之间的关系，涉及遮罩图像、边界框检测和迁移学习等方法。通过使用多种解释性和敏感性分析方法，我们调查了模型如何与其在病变区域上的注意力对齐，并探讨这种对齐程度如何与精度、召回率和F1分数相关联。结果显示，关注病变区域更密切的模型实现了更好的诊断性能，这表明可解释的人工智能技术在未来医学诊断中的潜力。这项研究为开发更加准确和可信的黑色素瘤分类模型奠定了基础。
URL
https://arxiv.org/abs/2601.00355
PDF
-
Attention to Detail: Global-Local Attention for High-Resolution AI-Generated Image Detection
2026-01-01 00:00:07Lawrence HanAbstract
The rapid development of generative AI has made AI-generated images increasingly realistic and high-resolution. Most AI-generated image detection architectures typically downsample images before inputting them into models, risking the loss of fine-grained details. This paper presents GLASS (Global-Local Attention with Stratified Sampling), an architecture that combines a globally resized view with multiple randomly sampled local crops. These crops are original-resolution regions efficiently selected through spatially stratified sampling and aggregated using attention-based scoring. GLASS can be integrated into vision models to leverage both global and local information in images of any size. Vision Transformer, ResNet, and ConvNeXt models are used as backbones, and experiments show that GLASS outperforms standard transfer learning by achieving higher predictive performance within feasible computational constraints.
Abstract (translated)
生成式人工智能的快速发展使得AI生成的图像越来越逼真且分辨率更高。大多数用于检测AI生成图像的架构通常会在将图像输入模型之前对其进行降采样，这会带来细节损失的风险。本文提出了一种名为GLASS（分层抽样全局-局部注意力）的架构，该架构结合了全局重缩放视图与多个随机抽取的局部裁剪块。这些裁剪块是在空间分层抽样的基础上高效选择的原始分辨率区域，并通过基于注意力的评分机制进行聚合。GLASS可以集成到视觉模型中，以利用任何大小图像中的全局和局部信息。实验使用了Vision Transformer、ResNet 和 ConvNeXt 模型作为骨干网络，结果表明，在可行的计算约束下，GLASS能够超越标准迁移学习方法，实现更高的预测性能。
URL
https://arxiv.org/abs/2601.00141
PDF
-
Deep Learning Approach for the Diagnosis of Pediatric Pneumonia Using Chest X-ray Imaging
2025-12-31 00:07:06Fatemeh Hosseinabadi, Mohammad Mojtaba RohaniAbstract
Pediatric pneumonia remains a leading cause of morbidity and mortality in children worldwide. Timely and accurate diagnosis is critical but often challenged by limited radiological expertise and the physiological and procedural complexity of pediatric imaging. This study investigates the performance of state-of-the-art convolutional neural network (CNN) architectures ResNetRS, RegNet, and EfficientNetV2 using transfer learning for the automated classification of pediatric chest Xray images as either pneumonia or normal.A curated subset of 1,000 chest X-ray images was extracted from a publicly available dataset originally comprising 5,856 pediatric images. All images were preprocessed and labeled for binary classification. Each model was fine-tuned using pretrained ImageNet weights and evaluated based on accuracy and sensitivity. RegNet achieved the highest classification performance with an accuracy of 92.4 and a sensitivity of 90.1, followed by ResNetRS (accuracy: 91.9, sensitivity: 89.3) and EfficientNetV2 (accuracy: 88.5, sensitivity: 88.1).
Abstract (translated)
儿科肺炎仍然是全球儿童发病率和死亡率的主要原因。及时且准确的诊断至关重要，但往往由于放射学专业知识有限以及儿科影像生理和操作复杂性而受到挑战。本研究调查了使用迁移学习的最先进的卷积神经网络（CNN）架构ResNetRS、RegNet和EfficientNetV2在自动分类儿童胸部X光图像为肺炎或正常情况下的性能。 从一个包含5,856张儿科影像数据集的公开可用数据集中，提取并精选出1,000张胸部X光片进行研究。所有图像均经过预处理，并进行了二元分类标签标注（即肺炎或正常）。每种模型使用预训练的ImageNet权重微调，并基于准确率和灵敏度进行评估。 结果显示，RegNet达到了最高的分类性能，其准确率为92.4%，灵敏度为90.1%；其次为ResNetRS，准确率为91.9%，灵敏度为89.3%；EfficientNetV2则分别为88.5%和88.1%。
URL
https://arxiv.org/abs/2601.00041
PDF
-
StressRoBERTa: Cross-Condition Transfer Learning from Depression, Anxiety, and PTSD to Stress Detection
2025-12-29 19:16:14Amal Alqahtani, Efsun Kayi, Mona DiabAbstract
The prevalence of chronic stress represents a significant public health concern, with social media platforms like Twitter serving as important venues for individuals to share their experiences. This paper introduces StressRoBERTa, a cross-condition transfer learning approach for automatic detection of self-reported chronic stress in English tweets. The investigation examines whether continual training on clinically related conditions (depression, anxiety, PTSD), disorders with high comorbidity with chronic stress, improves stress detection compared to general language models and broad mental health models. RoBERTa is continually trained on the Stress-SMHD corpus (108M words from users with self-reported diagnoses of depression, anxiety, and PTSD) and fine-tuned on the SMM4H 2022 Task 8 dataset. StressRoBERTa achieves 82% F1-score, outperforming the best shared task system (79% F1) by 3 percentage points. The results demonstrate that focused cross-condition transfer from stress-related disorders (+1% F1 over vanilla RoBERTa) provides stronger representations than general mental health training. Evaluation on Dreaddit (81% F1) further demonstrates transfer from clinical mental health contexts to situational stress discussions.
Abstract (translated)
慢性压力的普遍性代表了一个重要的公共健康问题，社交媒体平台如Twitter成为了个人分享其经历的重要场所。本文介绍了StressRoBERTa，这是一种用于自动检测英文推特中自我报告的慢性压力的跨条件迁移学习方法。研究探讨了在与临床相关状况（抑郁症、焦虑症、PTSD）上进行持续训练是否能提高慢性压力检测的效果，而这些状况具有较高的共病性，并且优于一般的语言模型和广泛的心理健康模型。 RoBERTa模型在Stress-SMHD语料库（来自自述患有抑郁症、焦虑症和PTSD的用户，共计1.08亿单词）上进行持续训练，并在SMM4H 2022 Task 8数据集上进行了微调。StressRoBERTA达到了82%的F1值，比共享任务中的最佳系统（79% F1）高出3个百分点。结果显示，从与压力相关疾病的跨条件转移（相对于原始RoBERTa模型提升了1% F1），提供了比一般心理健康训练更强的表现力。 在Dreaddit数据集上的评估进一步表明了从临床心理健康背景到情境性压力讨论的迁移能力，F1值为81%。
URL
https://arxiv.org/abs/2512.23813
PDF
-
Le Cam Distortion: A Decision-Theoretic Framework for Robust Transfer Learning
2025-12-29 17:21:44Deniz AkdemirAbstract
Distribution shift is the defining challenge of real-world machine learning. The dominant paradigm--Unsupervised Domain Adaptation (UDA)--enforces feature invariance, aligning source and target representations via symmetric divergence minimization [Ganin et al., 2016]. We demonstrate that this approach is fundamentally flawed: when domains are unequally informative (e.g., high-quality vs degraded sensors), strict invariance necessitates information destruction, causing "negative transfer" that can be catastrophic in safety-critical applications [Wang et al., 2019]. We propose a decision-theoretic framework grounded in Le Cam's theory of statistical experiments [Le Cam, 1986], using constructive approximations to replace symmetric invariance with directional simulability. We introduce Le Cam Distortion, quantified by the Deficiency Distance $\delta(E_1, E_2)$, as a rigorous upper bound for transfer risk conditional on simulability. Our framework enables transfer without source degradation by learning a kernel that simulates the target from the source. Across five experiments (genomics, vision, reinforcement learning), Le Cam Distortion achieves: (1) near-perfect frequency estimation in HLA genomics (correlation $r=0.999$, matching classical methods), (2) zero source utility loss in CIFAR-10 image classification (81.2% accuracy preserved vs 34.7% drop for CycleGAN), and (3) safe policy transfer in RL control where invariance-based methods suffer catastrophic collapse. Le Cam Distortion provides the first principled framework for risk-controlled transfer learning in domains where negative transfer is unacceptable: medical imaging, autonomous systems, and precision medicine.
Abstract (translated)
分布迁移是现实世界机器学习面临的最大挑战。主导的方法——无监督领域适应（UDA）强制要求特征不变性，通过对称散度最小化来使源域和目标域的表示相匹配 [Ganin et al., 2016]。我们证明了这种方法本质上是有缺陷的：当领域信息量不等（如高质量传感器与退化的传感器之间），严格的不变性需要破坏某些信息，从而导致“负迁移”，这种现象在安全关键应用中可能会非常严重 [Wang et al., 2019]。我们提出了一种基于Le Cam统计实验理论的决策论框架[Le Cam, 1986]，通过构造性的近似方法将对称不变性替换为方向可模拟性。我们引入了Le Cam扭曲度量，并用匮乏距离$\delta(E_1, E_2)$进行量化，作为条件下的迁移风险严格上限。我们的框架允许在不损害源域性能的情况下实现迁移学习，通过学习一个能够从源域模拟目标域的核函数来完成这一过程。 在五项实验（基因组学、视觉和强化学习）中，Le Cam扭曲实现了以下成果： 1. 在HLA基因分型中的近乎完美的频率估计（相关系数$r=0.999$，与经典方法相当） 2. CIFAR-10图像分类任务中源域性能几乎无损失的保持（准确率从81.2%降到34.7%，而CycleGAN则下降到了56.5%） 3. 在RL控制中的安全策略迁移，在不变性方法会导致灾难性崩溃的情况下，Le Cam扭曲框架下可以实现稳定传输。 Le Cam扭曲为风险可控的学习迁移提供了第一个原则性的框架，特别适用于不允许负迁移的领域：医学成像、自主系统和精准医疗。
URL
https://arxiv.org/abs/2512.23617
PDF
