Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks99 from public data
- Plug-in Diffusion Model for Sequential Recommendation112
A novel Plug-in Diffusion Model for Recommendation (PDRec) framework, which employs the diffusion model as a flexible plugin to jointly take full advantage of the diffusion-generating user preferences on all items to alleviate data sparsity.
- Modeling Disease Progression via Multisource Multitask Learners: A Case Study With Alzheimer’s Disease99
A novel and unified scheme to coregularize the prior knowledge of source consistency and temporal smoothness of chronic disease progression is proposed and theoretically proves that the proposed model is a linear model.
- 83
- Multi-Source Domain Adaptation for Visual Sentiment Classification79
A novel multi-source domain adaptation (MDA) method, termed Multi-source Sentiment Generative Adversarial Network (MSGAN), for visual sentiment classification that learns to find a unified sentiment latent space where data from both the source and target domains share a similar distribution.
- 76
- Multimodal Conditioned Diffusion Model for Recommendation72
A novel Multimodal Conditioned Diffusion Model for Recommendation (MCDRec), which tailors DM with two technical modules to model the high-order multimodal knowledge and bridges the insurmountable gap between the multi-modal content features and the collaborative signals.
- Heterogeneous Fusion of Semantic and Collaborative Information for Visually-Aware Food Recommendation70
Experimental results from two real-world datasets show that the visual features generated by PiNet better attend to the informative image regions, yielding superior performance.
- Cross-Silo Prototypical Calibration for Federated Learning with Non-IID Data69
A cross-silo prototypical calibration method (FedCSPC), which takes additional prototype information from the clients to learn a unified feature space on the server side and develops an augmented contrastive learning method to improve the robustness of the calibration.
- Towards online and personalized daily activity recognition, habit modeling, and anomaly detection for the solitary elderly through unobtrusive sensing65
A novel framework, named the online daily habit modeling and anomaly detection (ODHMAD) model, for the real-time personalized ADL recognition, habit modeling, and anomalies detection for the solitary elderly is presented.
- Triple Sequence Learning for Cross-domain Recommendation61
A novel framework, termed triple sequence learning for cross-domain recommendation (Tri-CDR), which jointly models the source, target, and mixed behavior sequences to highlight the global and target preference and precisely model the triple correlation in CDR.
- 54
- 41
- Self-organizing neural networks for universal learning and multimodal memory encoding40
This paper shows how a family of biologically-inspired self-organizing neural networks, known as fusion Adaptive Resonance Theory (fusion ART), may provide a viable approach to realizing the learning and memory functions.
- Learning Using Privileged Information for Food Recognition40
Experiments on two real-world datasets show that ATNet can be incorporated with any state-of-the-art CNN models to consistently improve their performance and is motivated by the paradigm of learning using privileged information (LUPI).
- Exploring False Hard Negative Sample in Cross-Domain Recommendation33
A novel model-agnostic Real Hard Negative Sampling (RealHNS) framework specially for cross- domain recommendation (CDR), which aims to discover the false and refine the real from all HNS via both general and cross-domain real hard negative sample selectors.
- Used Car Price Prediction Based on the Iterative Framework of XGBoost+LightGBM32
It is verified that the iterative framework proposed in this paper can be applied to other models and greatly improve the prediction performance of other models.
- Negative Sampling in Recommendation: A Survey and Future Directions29
This survey discusses existing user feedback, the critical role of negative sampling and the optimization objectives in RS, and thoroughly analyze challenges that consistently impede its progress, including an extensive literature review on the existing negative sampling strategies in RS.
- Cross-Training with Prototypical Distillation for improving the generalization of Federated Learning26
- 25
- Clustering-based Curriculum Construction for Sample-Balanced Federated Learning25
A Clustering-based curriculum construction method to rank the complexity of instances, and a Federation curriculum learning algorithm (FedAC) that assigns different weights to training samples of different complexity to take full advantage of the valuable learning knowledge from a noisy and uneven-quality data is proposed.
- Cross-modal learning using privileged information for long-tailed image classification24
A cross-modal enhanced method for long-tailed image classification, referred to as CMLTNet, which improves the learning of intraclass similarity of tail-class representations by cross-modal alignment and captures the difference between the head and tail classes in semantic space by cross-modal inference.
- 23
- Anything to Glyph: Artistic Font Synthesis via Text-to-Image Diffusion Model23
This paper proposes a novel method for generating creative artistic fonts using a pre-trained text-to-image diffusion model that can preserve the object’s original shape while constraining its position and introduces a novel heatmap-based weak position constraint method.
- 20
- Prompt Learning with Cross-Modal Feature Alignment for Visual Domain Adaptation19
Aiming at the problem of knowledge forgetting during ”fine-tuning”, an image classification algorithm (PMHANet) integrating a pre-trained large-scale model and heterogeneous feature alignment is proposed, providing a cross-modal knowledge transfer paradigm for multimodal pre-training of large models.
- Comparative Study of Adversarial Training Methods for Long-tailed Classification19
Experimental results on two benchmarking datasets show that a combination of statistical perturbations and hybrid optimization achieves a promising performance, and the gradient-based method typically improves the performance of both the head and tail classes.
- Comparative Study of Adversarial Training Methods for Cold-Start Recommendation19
Experimental results on two benchmarking datasets show that existing state-of-the-art algorithms can be further improved by a proper pairing of the key components as listed in the proposed taxonomy for adversarial training.
- Salience-aware adaptive resonance theory for large-scale sparse data clustering19
This paper presents a class of self-organizing neural networks, called the salience-aware adaptive resonance theory (SA-ART) model, which extends Fuzzy ART with measures for cluster-wise salient feature modeling to alleviate the side-effect of noisy features incurred by high dimensionality.
- 19
- Cross-Modal Content Inference and Feature Enrichment for Cold-Start Recommendation18
A recommendation framework, named CIERec, which exploits the multi-modal information to improve its cold-start recommendation performance and can consistently achieve significant improvements with different conventional visually-aware backbones, which verifies its universality and effectiveness.
- 18
- 16
- 16
- 16
- MLFont: Few-Shot Chinese Font Generation via Deep Meta-Learning16
A novel deep meta-learning-based font generation method (MLFont) for few-shot Chinese font generation, which leverages existing fonts to improve the generalization capability of the model for new fonts.
- 16
- A dual scheduling framework for task and resource allocation in clouds using deep reinforcement learning15
Experimental results show that the proposed intelligent dual scheduling framework is effective in ensuring higher task success rates, improving resource utilization, shortening average task response time, and significantly reducing the cost of renting VM instances compared to other baseline solutions.
- Modeling Event-level Causal Representation for Video Classification15
An Event-level Causal Representation Learning model for the spatio-temporal modeling of both the in-frame object interactions and their cross-frame temporal correlations is presented.
- 15
- Causal Inference over Visual-Semantic-Aligned Graph for Image Classification14
Experimental results from two benchmarking datasets show that VSCNet effectively builds the visual-semantic associations from images and leads to better performance than the state-of-the-art methods with enriched predictive information.
- Class-level Structural Relation Modeling and Smoothing for Visual Representation Learning14
Experiments demonstrate the effectiveness of structured knowledge modelling for enhanced representation learning and show that CSRMS can be incorporated with any state-of-the-art visual representation learning models for performance gains.
- Unsupervised Contrastive Masking for Visual Haze Classification13
An unsupervised contrastive masking (UCM) algorithm to segment the haze regions without any supervision is presented, and a dual-channel model-agnostic framework, termed magnifier neural network (MagNet), is developed to effectively use the segmented haze regions to enhance the learning of haze features by conventional deep learning models.
- 13
- FashionDPO:Fine-tune Fashion Outfit Generation Model using Direct Preference Optimization11
This framework aims to provide a general fine-tuning approach to fashion generative models, refining a pre-trained fashion outfit generation model using automatically generated feedback, without the need to design a task-specific reward function.
- DeepMCGCN: Multi-channel Deep Graph Neural Networks10
The DeepMCGCN method is evaluated on three real-world datasets, and the experimental results show that its node classification performance outperforms that of single-channel GCN and other benchmark models, which improves the modeling and representation capabilities of the model.
- HR-PrGAN: High-resolution story visualization with progressive generative adversarial networks10
A progressive adversarial learning algorithm, termed HR-PrGAN, to achieve high-resolution image sequences with rich details by decomposing the problem of generating into multiple stages and outperforms state-of-the-art methods in terms of image anti-deformation, feature synthesis and human perception based image quality evaluation.
- Disentangled Representations and Hierarchical Refinement of Multi-Granularity Features for Text-to-Image Synthesis10
A Hierarchical Disentangled Representations Generative Adversarial Networks (HDR-GAN) is proposed to generate photo-realistic images by explicitly disentangling and individually modeling the factors of semantics in the image.
- Towards Age-friendly E-commerce Through Crowd-Improved Speech Recognition, Multimodal Search, and Personalized Speech Feedback10
An age-friendly system to integrate three assistive techniques to facilitate the elderly's browsing of products in E-commerce platforms, including the crowd-improved speech recognition, the multimodal search, and the personalized speech feedback.
- Line graph contrastive learning for node classification9
Results validate the effectiveness and superiority of LineGCL in capturing and understanding global structural information, surpassing the limitations of baseline methods in graph contrastive learning.
- Exploring Web Images to Enhance Skin Disease Analysis Under A Computer Vision Framework9
A novel transfer learning model is presented, which is able to incorporate external knowledge obtained from the rich and relevant Web images contributed by grassroots, and is easily repeatable by other researchers and extendable to other disease types.
- 8
- DreamFont3D: Personalized Text-to-3D Artistic Font Generation8
To enhance the recognizability of 3D fonts, the multi-view mask constraint (MC) is proposed to optimize the differentiable 3D representation while preserving the font structure, and a progressive mask weighting module is presented to ensure a trade-off between the text-guided stylization of font effects and the mask-guided preservation of font structure.
- Causal inference for out‐of‐distribution recognition via sample balancing8
A balanced causal learning framework (BCL) is proposed, starting from how to divide the dataset in a balanced way and the balance of training after the division, which automatically generates fine‐grained balanced data partitions in an unsupervised manner and balances the training difficulty of different classes, thereby enhancing the generalisation ability of models in different environments.
- Class-aware Convolution and Attentive Aggregation for Image Classification7
A class-aware convolution and attentive aggregation framework that improves the effectiveness of representation learning and reduces the influence of irrelevant background and improves the performance of state-of-the-art backbones and enhances feature extraction robustness.
- 7
- 7
- Adaptive Resonance Theory (ART) for Social Media Analytics6
This chapter presents the ART-based clustering algorithms for social media analytics in detail, and introduces Fuzzy ART and its clustering mechanisms, which provides a deep understanding of the base model that is used and extended for handling the social media clustering challenges.
- Usability Analysis of the Novel Functions to Assist the Senior Customers in Online Shopping6
This paper extends the previous e-commerce website design with novel assistance functions including multimodal search and personalized speech feedback and evaluated the usability of these functions through a phenomenography based qualitative study.
- Novel Functional Technologies for Age-Friendly E-commerce6
This paper designs a multimodal product search function, and develops a product reputation function to provide an objective evaluation of products’ quality, which helps the elderly filter out low-quality products while saves their energy in product comparison.
- 6
- Fast training of Support Vector Machines using error-center-based optimization6
A new algorithm for Support Vector Machine (SVM) training, which trains a machine based on the cluster centers of errors caused by the current machine, which scales almost linear with training set size and thus may be applied to much larger training sets.
- 5
- Sequential selection and calibration of video frames for 3D outdoor scene reconstruction5
A sparse multi‐view images‐based 3D scene reconstruction framework (SMSR), which divides the scene reconstruction task into three stages: initial prediction, refinement, and fusion stage, and achieves superior performance in the scene understanding and reconstruction.
- 4
- Learning to Fuse Residual and Conditional Information for Video Compression and Reconstruction4
Experimental results from two benchmark datasets show that the proposed video compression method can effectively remove redundancy between video frames and reconstruct video frames with low distortion to achieve better than state-of-the-art (SOTA) performance.
- 4
- 3
- 3
- TransCotANet: A Lung Field Image Segmentation Network with Multidimensional Global Feature Dynamic Aggregation3
A new network architecture called TransCotANet is proposed, based on the U-Net architecture with convolutional neural networks (CNNs) as the backbone and extracts global semantic information through symmetric cross-layer connections in the encoder structure.
- Online Multimodal Co-indexing and Retrieval of Social Media Data3
This chapter presents a study on using the Online Multimodal Co-indexing Adaptive Resonance Theory (OMC-ART) for an effective and efficient indexing and retrieval of social media data.
- 2
- 2
- A Graph Representation Learning Framework Predicting Potential Multivariate Interactions2
Ahyperlink prediction framework for predicting multivariate interactions based on graph representation learning is proposed to solve the above problems, and the framework is extended to directed hyperlink prediction (e.g., directed metabolic reaction networks).
- VAFA: A Visually-Aware Food Analysis System for Socially-Engaged Diet Management2
A visually-aware food analysis (VAFA) system for socially-engaged diet management that can recommend individualized recipes and detect social communities that share similar dietary appetites for users by profiling users’ eating habits is demonstrated.
- Socially-Enriched Multimedia Data Co-clustering2
This chapter explains how to use the Generalized Heterogeneous Fusion Adaptive Resonance Theory (GHF-ART) for clustering large-scale web multimedia documents and reveals an adaptive method for effective fusion of the multimodal features.
- 2
- 1
- InstantAS: Minimum Coverage Sampling for Arbitrary-Size Image Generation1
This paper presents the InstantAS method for arbitrary-size image generation, and proposes the dynamic semantic guided cross-attention method, allowing for the control of different regions using different semantics.
- Unifying Visual and Semantic Feature Spaces with Diffusion Models for Enhanced Cross-Modal Alignment1
- Unsupervised Segmentation of Haze Regions as Hard Attention for Haze Classification1
A hard attention infused network (HAINet) is proposed for haze classification, consisting of an unsupervised segmentation module (USM) and a hybrid information fusion module (HIF) that reduces the impact of background noise in haze images and consistently improves the classification performance.
- Clustering and Its Extensions in the Social Media Domain1
This chapter summarizes existing clustering and related approaches for the identified challenges as described in Sect.
- Facades structure detection by geometric moment1
A novel method for extracting facades structure from real-world pictures by using local geometric moment, which has advantages of easy-to-implement, low computational cost, and robustness to noises, such as uneven illumination, shadow, and shade from other objects.
- –
- –
- –
- –
- –
- Investigating the mechanisms of malignant progression in colorectal cancer using weighted gene co-expression network analysis and machine learning–
The TFAP4-CDK1 axis is established as a critical driver of malignant progression in CRC and Targeting this pathway could lead to the development of novel interventions for CRC.
- –
- LLM-Enabled Style and Content Regularization for Personalized Text-to-Image Generation–
The style refinement strategy leverages the semantic information of visual reasoning prompts and reference images to optimize style embeddings, allowing a more precise and consistent representation of style information.
- –
- –
- TP-GCL: graph contrastive learning from the tensor perspective–
This work proposes a novel graph contrastive learning method, TP-GCL, based on a tensor perspective, to overcome the limitations of traditional GNNs in modeling complex structures and addressing the issue of sparse labels in graph data.
- –
- –
- Personalized Web Image Organization–
This chapter presents using the Probabilistic ART with user preference architecture, which groups images of similar semantics together and simultaneously mines the key tags/topics of individual clusters to give users direct control of the generated clusters.
- –
- A Fourier model of periodic dynamic speech balloons–
A Fourier model to represent the dynamics of a speech balloon is proposed and can be used to generate various dynamic speech balloons with different styles.
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.