arXiv subjects

C. -C. Jay Kuo

Publications and source records attributed to C. -C. Jay Kuo.

At least 19 recordsLinked to original sources

arXiv2026

3D-PLOT-LLM: Part-Level Object Tokens for 3D Large Language Models

3D multimodal large language models (3D MLLMs) describe a 3D object as a whole but cannot address, name, or reason about its parts. Prior part-aware attempts add segmentation decoders, heavier 3D encoders, or bounding-box grammars at substantial parameter cost. We take a fundamentally different path: we reorganize the input token stream so that parts become directly addressable through the LLM's own vocabulary. Our model, 3D-PLOT-LLM, partitions the frozen point encoder's patches into K locally coherent regions and inserts, before each region's patch tokens, a learnable per-region marker and a reserved vocabulary token ; a Marker-Space Refinement (MSR) module then conditions each marker on its region's spatial statistics and adjacency neighbors. The model thus cites parts in its output and follows prompts that refer to parts by token, a capability absent from prior object-level 3D MLLMs. To probe this interface, we construct PartVerse-QA, a vocabulary-level part-QA benchmark adapted from PartVerse mesh annotations (77K training pairs and 588 held-out queries on disjoint object splits), on which 3D-PLOT-LLM reaches caption-to-slots Jaccard 0.459 and Exact-match 13.78%, with a slot-to-caption GPT-4o judge of 44.68. On the 3DCoMPaT-GrIn part-aware grounded description benchmark, 3D-PLOT-LLM outperforms PointLLM, Kestrel, PARIS3D, and SegPoint on every text-output metric, and ShapeLLM on 3 of 4, with up to +3.03 GPT-4o judge over PointLLM. On Objaverse whole-object captioning, adding PartVerse-QA at Stage 2 yields +0.65 SBERT and +1.85 GPT-4o over PointLLM, and tops PointLLM-PiSA on 4 of 5 traditional metrics (SBERT, SimCSE, BLEU-1, METEOR) despite targeting a different (part-grounded) objective. All with under 1M new trainable parameters on a frozen point encoder, an order of magnitude below prior part-aware 3D MLLMs, and no segmentation decoder or bounding-box head.

cs.CV↗

arXiv2026

Interpretable and backpropagation-free Green Learning for efficient multi-task echocardiographic segmentation and classification

Echocardiography is a cornerstone for managing heart failure (HF), with Left Ventricular Ejection Fraction (LVEF) being a critical metric for guiding therapy. However, manual LVEF assessment suffers from high inter-observer variability, while existing Deep Learning (DL) models are often computationally intensive and data-hungry "black boxes" that impede clinical trust and adoption. Here, we propose a backpropagation-free multi-task Green Learning (MTGL) framework that performs simultaneous Left Ventricle (LV) segmentation and LVEF classification. Our framework integrates an unsupervised VoxelHop encoder for hierarchical spatio-temporal feature extraction with a multi-level regression decoder and an XG-Boost classifier. On the EchoNet-Dynamic dataset, our MTGL model achieves state-of-the-art classification and segmentation performance, attaining a classification accuracy of 94.3% and a Dice Similarity Coefficient (DSC) of 0.912, significantly outperforming several advanced 3D DL models. Crucially, our model achieves this with over an order of magnitude fewer parameters, demonstrating exceptional computational efficiency. This work demonstrates that the GL paradigm can deliver highly accurate, efficient, and interpretable solutions for complex medical image analysis, paving the way for more sustainable and trustworthy artificial intelligence in clinical practice.

eess.IV↗

arXiv2026

HCL-FF: Hierarchical and Contrastive Learning for Forward-Forward Algorithm

Deep neural networks trained with backpropagation have achieved outstanding performance in vision tasks but remain biologically implausible, computationally demanding, and difficult to interpret. The Forward-Forward (FF) algorithm offers a promising alternative by training each layer independently through local goodness objectives. However, its purely local optimization lacks hierarchical coordination across layers, and the decoupling of goodness from features leaves the representations unconstrained and semantically ambiguous. We propose a Hierarchical and Contrastive Learning FF framework (HCL-FF) to address these limitations. HCL-FF introduces (1) a coarse-to-fine hierarchical learning strategy that guides representations from low-level cues to high-level semantics, and (2) a supervised contrastive objective that enforces class-discriminative alignment after goodness decoupling. Experiments on CIFAR-10, CIFAR-100, and Tiny-ImageNet demonstrate that HCL-FF achieves new state-of-the-art performance among FF-based methods, with notable accuracy gains of +5.46%, +17.00%, and +12.51%, respectively.

cs.CV↗

arXiv2026

Automatic Modulation Classification via Green Machine Learning

In this work, we propose an interpretable, robust, and lightweight machine learning method for automatic modulation classification (AMC) under dynamic and noisy channel conditions. It is called green automatic modulation classification (GAMC) and targets edge artificial intelligence (AI) with low computational complexity and a small model size. GAMC operates in four stages. First, raw received I/Q signals are transformed into multi-domain representations, including constellation diagrams and spatio-temporal graphs. Second, we extract a comprehensive set of statistical and topological features from time-series signals, constellation diagrams, and graphs. Third, a supervised feature learning process leverages label guidance to project high-dimensional features into robust, discriminative low-dimensional ones. Finally, a context-aware Signal-to-Noise Ratio (SNR) soft routing mechanism ensembles predictions from downstream classifiers. Experimental results show that GAMC effectively mitigates domain shifts caused by high noise. It strikes a good balance between accuracy and efficiency, reducing the number of model parameters by $50\%$, operating at $3\%$ to $42\%$ of the computational cost of lightweight deep learning models, and maintaining higher accuracy in various SNRs.

eess.SP↗

arXiv2026

G-AMC: A Green Automatic Modulation Classification Method

In this work, we propose an efficient and transparent green learning pipeline to address the automatic modulation classification (AMC) problem. This pipeline aims to enable receivers to blindly identify the modulation modes of the incoming signals in a computationally efficient way with a small model size. Our method includes the following steps. First, the input signal is transformed into a precise representation through the sparse coding method. Second, various features are extracted from the sparse coding representation with the statistics from the input signal. Third, the classification subspace is hierarchically partitioned with a tree structure to achieve a lightweight model size with good prediction accuracy. The experimental results demonstrate the effectiveness and efficiency in classifying the modulated features and representation of received signals. Compared to lightweight deep learning models, the number of model parameters is reduced by \textbf{41\%}, while the usage of Floating Point Operations (FLOPs) is only $\mathcal{O}(10^{-4})$ of the blind waveform recognition without pre-arranged knowledge of incoming waveforms.

eess.SP↗

arXiv2026

MemRoPE: Training-Free Infinite Video Generation via Evolving Memory Tokens

Autoregressive diffusion enables real-time frame streaming, yet existing sliding-window caches discard past context, causing fidelity degradation, identity drift, and motion stagnation over long horizons. Current approaches preserve a fixed set of early tokens as attention sinks, but this static anchor cannot reflect the evolving content of a growing video. We introduce MemRoPE, a training-free framework with two co-designed components. Memory Tokens continuously compress all past keys into dual long-term and short-term streams via exponential moving averages, maintaining both global identity and recent dynamics within a fixed-size cache. Online RoPE Indexing caches unrotated keys and applies positional embeddings dynamically at attention time, ensuring the aggregation is free of conflicting positional phases. These two mechanisms are mutually enabling: positional decoupling makes temporal aggregation well-defined, while aggregation makes fixed-size caching viable for unbounded generation. Extensive experiments validate that MemRoPE outperforms existing methods in temporal coherence, visual fidelity, and subject consistency across minute- to hour-scale generation.

cs.CV↗

arXiv2026

GreenPhase: A Green Learning Approach for Earthquake Phase Picking

Earthquake detection and seismic phase picking are fundamental yet challenging tasks in seismology due to low signal-to-noise ratios, waveform variability, and overlapping events. Recent deep-learning models achieve strong results but rely on large datasets and heavy backpropagation training, raising concerns over efficiency, interpretability, and sustainability. We propose GreenPhase, a multi-resolution, feed-forward, and mathematically interpretable model based on the Green Learning framework. GreenPhase comprises three resolution levels, each integrating unsupervised representation learning, supervised feature learning, and decision learning. Its feed-forward design eliminates backpropagation, enabling independent module optimization with stable training and clear interpretability. Predictions are refined from coarse to fine resolutions while computation is restricted to candidate regions. On the Stanford Earthquake Dataset (STEAD), GreenPhase achieves excellent performance with F1 scores of 1.0 for detection, 0.98 for P-wave picking, and 0.96 for S-wave picking. This is accomplished while reducing the computational cost (FLOPs) for inference by approximately 83% compared to state-of-the-art models. These results demonstrate that the proposed model provides an efficient, interpretable, and sustainable alternative for large-scale seismic monitoring.

physics.geo-ph↗

arXiv2026

A Statistical Approach for Modeling Irregular Multivariate Time Series with Missing Observations

Irregular multivariate time series with missing values present significant challenges for predictive modeling in domains such as healthcare. While deep learning approaches often focus on temporal interpolation or complex architectures to handle irregularities, we propose a simpler yet effective alternative: extracting time-agnostic summary statistics to eliminate the temporal axis. Our method computes four key features per variable-mean and standard deviation of observed values, as well as the mean and variability of changes between consecutive observations to create a fixed-dimensional representation. These features are then utilized with standard classifiers, such as logistic regression and XGBoost. Evaluated on four biomedical datasets (PhysioNet Challenge 2012, 2019, PAMAP2, and MIMIC-III), our approach achieves state-of-the-art performance, surpassing recent transformer and graph-based models by 0.5-1.7% in AUROC/AUPRC and 1.1-1.7% in accuracy/F1-score, while reducing computational complexity. Ablation studies demonstrate that feature extraction-not classifier choice-drives performance gains, and our summary statistics outperform raw/imputed input in most benchmarks. In particular, we identify scenarios where missing patterns themselves encode predictive signals, as in sepsis prediction (PhysioNet, 2019), where missing indicators alone can achieve 94.2% AUROC with XGBoost, only 1.6% lower than using original raw data as input. Our results challenge the necessity of complex temporal modeling when task objectives permit time-agnostic representations, providing an efficient and interpretable solution for irregular time series classification.

cs.LG↗

arXiv2026

A Green Learning Approach to LDCT Image Restoration

This work proposes a green learning (GL) approach to restore medical images. Without loss of generality, we use low-dose computed tomography (LDCT) images as examples. LDCT images are susceptible to noise and artifacts, where the imaging process introduces distortion. LDCT image restoration is an important preprocessing step for further medical analysis. Deep learning (DL) methods have been developed to solve this problem. We examine an alternative solution using the Green Learning (GL) methodology. The new restoration method is characterized by mathematical transparency, computational and memory efficiency, and high performance. Experiments show that our GL method offers state-of-the-art restoration performance at a smaller model size and with lower inference complexity.

cs.CV↗

arXiv2026

Cross-Modal Fine-Tuning of 3D Convolutional Foundation Models for ADHD Classification with Low-Rank Adaptation

Early diagnosis of attention-deficit/hyperactivity disorder (ADHD) in children plays a crucial role in improving outcomes in education and mental health. Diagnosing ADHD using neuroimaging data, however, remains challenging due to heterogeneous presentations and overlapping symptoms with other conditions. To address this, we propose a novel parameter-efficient transfer learning approach that adapts a large-scale 3D convolutional foundation model, pre-trained on CT images, to an MRI-based ADHD classification task. Our method introduces Low-Rank Adaptation (LoRA) in 3D by factorizing 3D convolutional kernels into 2D low-rank updates, dramatically reducing trainable parameters while achieving superior performance. In a five-fold cross-validated evaluation on a public diffusion MRI database, our 3D LoRA fine-tuning strategy achieved state-of-the-art results, with one model variant reaching 71.9% accuracy and another attaining an AUC of 0.716. Both variants use only 1.64 million trainable parameters (over 113x fewer than a fully fine-tuned foundation model). Our results represent one of the first successful cross-modal (CT-to-MRI) adaptations of a foundation model in neuroimaging, establishing a new benchmark for ADHD classification while greatly improving efficiency.

eess.IV↗

arXiv2025

Descrip3D: Enhancing Large Language Model-based 3D Scene Understanding with Object-Level Text Descriptions

Understanding 3D scenes goes beyond simply recognizing objects; it requires reasoning about the spatial and semantic relationships between them. Current 3D scene-language models often struggle with this relational understanding, particularly when visual embeddings alone do not adequately convey the roles and interactions of objects. In this paper, we introduce Descrip3D, a novel and powerful framework that explicitly encodes the relationships between objects using natural language. Unlike previous methods that rely only on 2D and 3D embeddings, Descrip3D enhances each object with a textual description that captures both its intrinsic attributes and contextual relationships. These relational cues are incorporated into the model through a dual-level integration: embedding fusion and prompt-level injection. This allows for unified reasoning across various tasks such as grounding, captioning, and question answering, all without the need for task-specific heads or additional supervision. When evaluated on five benchmark datasets, including ScanRefer, Multi3DRefer, ScanQA, SQA3D, and Scan2Cap, Descrip3D consistently outperforms strong baseline models, demonstrating the effectiveness of language-guided relational representation for understanding complex indoor scenes. Our code and data are publicly available at https://github.com/jintangxue/Descrip3D.

cs.CV↗

arXiv2025

LG-NuSegHop: A Local-to-Global Self-Supervised Pipeline For Nuclei Instance Segmentation

Nuclei segmentation is the cornerstone task in histology image reading, shedding light on the underlying molecular patterns and leading to disease or cancer diagnosis. Yet, it is a laborious task that requires expertise from trained physicians. The large nuclei variability across different organ tissues and acquisition processes challenges the automation of this task. On the other hand, data annotations are expensive to obtain, and thus, Deep Learning (DL) models are challenged to generalize to unseen organs or different domains. This work proposes Local-to-Global NuSegHop (LG-NuSegHop), a self-supervised pipeline developed on prior knowledge of the problem and molecular biology. There are three distinct modules: (1) a set of local processing operations to generate a pseudolabel, (2) NuSegHop a novel data-driven feature extraction model and (3) a set of global operations to post-process the predictions of NuSegHop. Notably, even though the proposed pipeline uses { no manually annotated training data} or domain adaptation, it maintains a good generalization performance on other datasets. Experiments in three publicly available datasets show that our method outperforms other self-supervised and weakly supervised methods while having a competitive standing among fully supervised methods. Remarkably, every module within LG-NuSegHop is transparent and explainable to physicians.

eess.IV↗

arXiv2025

GUSL-Dehaze: A Green U-Shaped Learning Approach to Image Dehazing

Image dehazing is a restoration task that aims to recover a clear image from a single hazy input. Traditional approaches rely on statistical priors and the physics-based atmospheric scattering model to reconstruct the haze-free image. While recent state-of-the-art methods are predominantly based on deep learning architectures, these models often involve high computational costs and large parameter sizes, making them unsuitable for resource-constrained devices. In this work, we propose GUSL-Dehaze, a Green U-Shaped Learning approach to image dehazing. Our method integrates a physics-based model with a green learning (GL) framework, offering a lightweight, transparent alternative to conventional deep learning techniques. Unlike neural network-based solutions, GUSL-Dehaze completely avoids deep learning. Instead, we begin with an initial dehazing step using a modified Dark Channel Prior (DCP), which is followed by a green learning pipeline implemented through a U-shaped architecture. This architecture employs unsupervised representation learning for effective feature extraction, together with feature-engineering techniques such as the Relevant Feature Test (RFT) and the Least-Squares Normal Transform (LNT) to maintain a compact model size. Finally, the dehazed image is obtained via a transparent supervised learning strategy. GUSL-Dehaze significantly reduces parameter count while ensuring mathematical interpretability and achieving performance on par with state-of-the-art deep learning models.

eess.IV↗

arXiv2025

Green Learning for STAR-RIS mmWave Systems with Implicit CSI

In this paper, a green learning (GL)-based precoding framework is proposed for simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS)-aided millimeter-wave (mmWave) MIMO broadcasting systems. Motivated by the growing emphasis on environmental sustainability in future 6G networks, this work adopts a broadcasting transmission architecture for scenarios where multiple users share identical information, improving spectral efficiency and reducing redundant transmissions and power consumption. Different from conventional optimization methods, such as block coordinate descent (BCD) that require perfect channel state information (CSI) and iterative computation, the proposed GL framework operates directly on received uplink pilot signals without explicit CSI estimation. Unlike deep learning (DL) approaches that require CSI-based labels for training, the proposed GL approach also avoids deep neural networks and backpropagation, leading to a more lightweight design. Although the proposed GL framework is trained with supervision generated by BCD under full CSI, inference is performed in a fully CSI-free manner. The proposed GL integrates subspace approximation with adjusted bias (Saab), relevant feature test (RFT)-based supervised feature selection, and eXtreme gradient boosting (XGBoost)-based decision learning to jointly predict the STAR-RIS coefficients and transmit precoder. Simulation results show that the proposed GL approach achieves competitive spectral efficiency compared to BCD and DL-based models, while reducing floating-point operations (FLOPs) by over four orders of magnitude. These advantages make the proposed GL approach highly suitable for real-time deployment in energy- and hardware-constrained broadcasting scenarios.

eess.SP↗

arXiv2025

GUSL: A Novel and Efficient Machine Learning Model for Prostate Segmentation on MRI

Prostate and zonal segmentation is a crucial step for clinical diagnosis of prostate cancer (PCa). Computer-aided diagnosis tools for prostate segmentation are based on the deep learning (DL) paradigm. However, deep neural networks are perceived as "black-box" solutions by physicians, thus making them less practical for deployment in the clinical setting. In this paper, we introduce a feed-forward machine learning model, named Green U-shaped Learning (GUSL), suitable for medical image segmentation without backpropagation. GUSL introduces a multi-layer regression scheme for coarse-to-fine segmentation. Its feature extraction is based on a linear model, which enables seamless interpretability during feature extraction. Also, GUSL introduces a mechanism for attention on the prostate boundaries, which is an error-prone region, by employing regression to refine the predictions through residue correction. In addition, a two-step pipeline approach is used to mitigate the class imbalance, an issue inherent in medical imaging problems. After conducting experiments on two publicly available datasets and one private dataset, in both prostate gland and zonal segmentation tasks, GUSL achieves state-of-the-art performance among other DL-based models. Notably, GUSL features a very energy-efficient pipeline, since it has a model size several times smaller and less complexity than the rest of the solutions. In all datasets, GUSL achieved a Dice Similarity Coefficient (DSC) performance greater than $0.9$ for gland segmentation. Considering also its lightweight model size and transparency in feature extraction, it offers a competitive and practical package for medical imaging applications.

eess.IV↗

arXiv2025

Deep-JGAC: End-to-End Deep Joint Geometry and Attribute Compression for Dense Colored Point Clouds

Colored point cloud becomes a fundamental representation in the realm of 3D vision. Effective Point Cloud Compression (PCC) is urgently needed due to huge amount of data. In this paper, we propose an end-to-end Deep Joint Geometry and Attribute point cloud Compression (Deep-JGAC) framework for dense colored point clouds, which exploits the correlation between the geometry and attribute for high compression efficiency. Firstly, we propose a flexible Deep-JGAC framework, where the geometry and attribute sub-encoders are compatible to either learning or non-learning based geometry and attribute encoders. Secondly, we propose an attribute-assisted deep geometry encoder that enhances the geometry latent representation with the help of attribute, where the geometry decoding remains unchanged. Moreover, Attribute Information Fusion Module (AIFM) is proposed to fuse attribute information in geometry coding. Thirdly, to solve the mismatch between the point cloud geometry and attribute caused by the geometry compression distortion, we present an optimized re-colorization module to attach the attribute to the geometrically distorted point cloud for attribute coding. It enhances the colorization and lowers the computational complexity. Extensive experimental results demonstrate that in terms of the geometry quality metric D1-PSNR, the proposed Deep-JGAC achieves an average of 82.96%, 36.46%, 41.72%, and 31.16% bit-rate reductions as compared to the state-of-the-art G-PCC, V-PCC, GRASP, and PCGCv2, respectively. In terms of perceptual joint quality metric MS-GraphSIM, the proposed Deep-JGAC achieves an average of 48.72%, 14.67%, and 57.14% bit-rate reductions compared to the G-PCC, V-PCC, and IT-DL-PCC, respectively. The encoding/decoding time costs are also reduced by 94.29%/24.70%, and 96.75%/91.02% on average as compared with the V-PCC and IT-DL-PCC.

cs.MM↗

arXiv2025

Green Video Camouflaged Object Detection

Camouflaged object detection (COD) aims to distinguish hidden objects embedded in an environment highly similar to the object. Conventional video-based COD (VCOD) methods explicitly extract motion cues or employ complex deep learning networks to handle the temporal information, which is limited by high complexity and unstable performance. In this work, we propose a green VCOD method named GreenVCOD. Built upon a green ICOD method, GreenVCOD uses long- and short-term temporal neighborhoods (TN) to capture joint spatial/temporal context information for decision refinement. Experimental results show that GreenVCOD offers competitive performance compared to state-of-the-art VCOD benchmarks.

cs.CV↗

arXiv2025

RadHop-Net: A Lightweight Radiomics-to-Error Regression for False Positive Reduction In MRI Prostate Cancer Detection

Clinically significant prostate cancer (csPCa) is a leading cause of cancer death in men, yet it has a high survival rate if diagnosed early. Bi-parametric MRI (bpMRI) reading has become a prominent screening test for csPCa. However, this process has a high false positive (FP) rate, incurring higher diagnostic costs and patient discomfort. This paper introduces RadHop-Net, a novel and lightweight CNN for FP reduction. The pipeline consists of two stages: Stage 1 employs data driven radiomics to extract candidate ROIs. In contrast, Stage 2 expands the receptive field about each ROI using RadHop-Net to compensate for the predicted error from Stage 1. Moreover, a novel loss function for regression problems is introduced to balance the influence between FPs and true positives (TPs). RadHop-Net is trained in a radiomics-to-error manner, thus decoupling from the common voxel-to-label approach. The proposed Stage 2 improves the average precision (AP) in lesion detection from 0.407 to 0.468 in the publicly available pi-cai dataset, also maintaining a significantly smaller model size than the state-of-the-art.

cs.CV↗