Author name cluster

Weiqi Zhang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

9 papers

1 author row

AAAI Conference 2026 Conference Paper

MedSpaformer: A Transferable Transformer with Multi-Granularity Token Sparsification for Medical Time Series Classification

Jiexia Ye
Weiqi Zhang
Ziyue Li
Jia Li
Fugee Tsung

Accurate medical time series (MedTS) classification is essential for effective clinical diagnosis, yet remains challenging due to complex multi-channel temporal dependencies, information redundancy, and label scarcity. While transformer-based models have shown promise in time series analysis, most are designed for forecasting tasks and fail to fully exploit the unique characteristics of MedTS. In this paper, we introduce MedSpaformer, a transformer-based framework tailored for MedTS classification. It incorporates a sparse token-based dual-attention mechanism that enables global context modeling and token sparsification, allowing dynamic feature refinement by focusing on informative tokens while reducing redundancy. This mechanism is integrated into a multi-granularity cross-channel encoding scheme to capture intra- and inter-granularity temporal dependencies and inter-channel correlations, enabling progressive refinement of task-relevant patterns in medical signals. The sparsification design allows our model to flexibly accommodate inputs with variable lengths and channel dimensions. We also introduce an adaptive label encoder to extract label semantics and address cross-dataset label space misalignment. Together, these components enhance the model’s transferability across heterogeneous medical datasets, which helps alleviate the challenge of label scarcity. Our model outperforms 13 baselines across 7 medical datasets under supervised learning. It also excels in few-shot learning and demonstrates zero-shot capability in both in-domain and cross-domain diagnostics. These results highlight MedSpaformer's robustness and its potential as a unified solution for MedTS classification across diverse settings.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

MaterialRefGS: Reflective Gaussian Splatting with Multi-view Consistent Material Inference

Wenyuan Zhang
Jimin Tang
Weiqi Zhang
Yi Fang
Yu-Shen Liu
Zhizhong Han

Modeling reflections from 2D images is essential for photorealistic rendering and novel view synthesis. Recent approaches enhance Gaussian primitives with reflection-related material attributes to enable physically based rendering (PBR) with Gaussian Splatting. However, the material inference often lacks sufficient constraints, especially under limited environment modeling, resulting in illumination aliasing and reduced generalization. In this work, we revisit the problem from a multi-view perspective and show that multi-view consistent material inference with more physically-based environment modeling is key to learning accurate reflections with Gaussian Splatting. To this end, we enforce 2D Gaussians to produce multi-view consistent material maps during deferred shading. We also track photometric variations across views to identify highly reflective regions, which serve as strong priors for reflection strength terms. To handle indirect illumination caused by inter-object occlusions, we further introduce an environment modeling strategy through ray tracing with 2DGS, enabling photorealistic rendering of indirect radiance. Experiments on widely used benchmarks show that our method faithfully recovers both illumination and geometry, achieving state-of-the-art rendering quality in novel views synthesis. Project Page: https: //wen-yuan-zhang. github. io/MaterialRefGS.

PDF Details

IJCAI Conference 2025 Conference Paper

MedualTime: A Dual-Adapter Language Model for Medical Time Series-Text Multimodal Learning

Jiexia Ye
Weiqi Zhang
Ziyue Li
Jia Li
Meng Zhao
Fugee Tsung

The recent rapid advancements in language models (LMs) have garnered attention in medical time series-text multimodal learning. However, existing contrastive learning-based and prompt-based LM approaches tend to be biased, often assigning a primary role to time series modality while treating text modality as secondary. We classify these approaches under a temporal-primary paradigm, which may overlook the unique and critical task-relevant information embedded in text modality like clinical reports, thus failing to fully leverage mutual benefits and complementarity of different modalities. To fill this gap, we propose a novel textual-temporal multimodal learning paradigm that enables either modality to serve as the primary while being enhanced by the other, thereby effectively capturing modality-specific information and fostering cross-modal interaction. In specific, we design MedualTime, a language model composed of dual adapters to implement temporal-primary and textual-primary modeling simultaneously. Within each adapter, lightweight adaptation tokens are injected into the top layers of LM to encourage high-level modality fusion. The shared LM pipeline by dual adapters not only achieves adapter alignment but also enables efficient fine-tuning, reducing computational resources. Empirically, MedualTime demonstrates superior performance on medical data, achieving notable improvements of 8% accuracy and 12% F1 in supervised settings. Furthermore, MedualTime's transferability is validated by few-shot transfer experiments from coarse-grained to fine-grained medical data.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Prototype-Guided Multimodal Relation Extraction based on Entity Attributes

Zefan Zhang
Weiqi Zhang
Yanhui Li
Tian Bai

Multimodal Relation Extraction (MRE) aims to predict relations between head and tail entities based on the context of sentence-image pairs. Most existing MRE methods progressively incorporate textual and visual inputs to dominate the learning process, assuming both contribute significantly to the task. However, the diverse visual appearances and text with ambiguous semantics contain less-informative contexts for the corresponding relation. To tackle these challenges, we highlight the importance of semantically invariant entity attributes that encompass fine-grained categories. Towards this, we propose a novel Prototype-Guided Multimodal Relation Extraction (PG-MRE) framework based on Entity Attributes. Specifically, we first generate detailed entity explanations using Large Language Models (LLMs) to supplement the attribute semantics. Then, the Attribute Prototype Module (APM) refines attribute categories and condenses scattered entity attribute features into cluster-level prototypes. Furthermore, prototype-aligned attribute features guide diverse visual appearance features to produce compact and distinctive multimodal representations in the Relation Prototype Module (RPM). Extensive experiments demonstrate that our method gains superior relation classification capability (especially in scenarios involving various unseen entities), achieving new state-of-the-art performances on MNRE dataset.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

DiffGS: Functional Gaussian Splatting Diffusion

Junsheng Zhou
Weiqi Zhang
Yu-Shen Liu

3D Gaussian Splatting (3DGS) has shown convincing performance in rendering speed and fidelity, yet the generation of Gaussian Splatting remains a challenge due to its discreteness and unstructured nature. In this work, we propose DiffGS, a general Gaussian generator based on latent diffusion models. DiffGS is a powerful and efficient 3D generative model which is capable of generating Gaussian primitives at arbitrary numbers for high-fidelity rendering with rasterization. The key insight is to represent Gaussian Splatting in a disentangled manner via three novel functions to model Gaussian probabilities, colors and transforms. Through the novel disentanglement of 3DGS, we represent the discrete and unstructured 3DGS with continuous Gaussian Splatting functions, where we then train a latent diffusion model with the target of generating these Gaussian Splatting functions both unconditionally and conditionally. Meanwhile, we introduce a discretization algorithm to extract Gaussians at arbitrary numbers from the generated functions via octree-guided sampling and optimization. We explore DiffGS for various tasks, including unconditional generation, conditional generation from text, image, and partial 3DGS, as well as Point-to-Gaussian generation. We believe that DiffGS provides a new direction for flexibly modeling and generating Gaussian Splatting. Project page: https: //junshengzhou. github. io/DiffGS.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

MultiPull: Detailing Signed Distance Functions by Pulling Multi-Level Queries at Multi-Step

Takeshi Noda
Chao Chen
Weiqi Zhang
Xinhai Liu
Yu-Shen Liu
Zhizhong Han

Reconstructing a continuous surface from a raw 3D point cloud is a challenging task. Latest methods employ supervised learning or pretrained priors to learn a signed distance function (SDF). However, neural networks tend to smooth local details due to the lack of ground truth signed distnaces or normals, which limits the performance of learning-based methods in reconstruction tasks. To resolve this issue, we propose a novel method, named MultiPull, to learn multi-scale implicit fields from raw point clouds to optimize accurate SDFs from coarse to fine. We achieve this by mapping 3D query points into a set of frequency features, which makes it possible to leverage multi-level features during optimization. Meanwhile, we introduce optimization constraints from the perspective of spatial distance and normal consistency, which play a key role in point cloud reconstruction based on multi-scale optimization strategies. Our experiments on widely used object and scene benchmarks demonstrate that our method outperforms the state-of-the-art methods in surface reconstruction.

PDF Details DOI

EAAI Journal 2023 Journal Article

TGM-Nets: A deep learning framework for enhanced forecasting of tumor growth by integrating imaging and modeling

Qijing Chen
Qi Ye
Weiqi Zhang
He Li
Xiaoning Zheng

Prediction and uncertainty quantification of tumor progression are vital in clinical practice, i. e. , disease prognosis and decision-making on treatment strategies. In this work, we propose TGM-Nets, a deep learning framework that combines bioimaging and tumor growth modeling (TGM) for enhanced prediction of tumor growth. This proposed framework, developed based on physics-informed neural networks (PINNs), is capable of integrating the TGM and sequential observations of tumor morphology for patient-specific prediction of tumor growth. The novelties of the design of TGM-Nets include the employment of Fourier layers to extract the features of the input images as well as the utilization of sequential learning and fine-tuning with physics for extrapolation to improve the prediction accuracy. The validity of TGM-Nets for tumor growth forecasting is verified by testing the model performance on synthetic and in-vitro datasets, respectively. Our results show that the TGM-Nets not only can track the growth rates of the mild and aggressive tumors but also capture their detailed morphological features within and outside the training domain. In particular, TGM-Nets can be used to predict the long time dynamics of tumor growth in mild and aggressive cases. Our results show that the parameters inferred from the TGM-Nets can be used for long-time prediction for up to 4 months with a maximum error of ∼ 4%. We also systematically study the effects of the number of training points and noisy data on the performance of TGM-Nets as well as quantify the uncertainty of the model predictions. We show that TGM-Nets can integrate the biomedical images to predict the growth of the in-vitro cultured pancreatic cancer cells and identify the associated growth rates, demonstrating the possibilities of using TGM-Nets in clinical practice. In summary, we propose a new deep learning model that combines imaging and TGM to improve the current approaches for predicting tumor growth and thus provide an advanced computational tool for patient-specific tumor prognosis.

Details DOI

IJCAI Conference 2022 Conference Paper

GRELEN: Multivariate Time Series Anomaly Detection from the Perspective of Graph Relational Learning

Weiqi Zhang
Chen Zhang
Fugee Tsung

System monitoring and anomaly detection is a crucial task in daily operation. With the rapid development of cyber-physical systems and IT systems, multiple sensors get involved to represent the system state from different perspectives, which inspires us to detect anomalies considering feature dependence relationship among sensors instead of focusing on individual sensor's behavior. In this paper, we propose a novel Graph Relational Learning Network (GReLeN) to detect multivariate time series anomalies from the perspective of between-sensor dependence relationship learning. Variational AutoEncoder (VAE) serves as the overall framework for feature extraction and system representation. Graph Neural Network (GNN) and stochastic graph relational learning strategy are also imposed to capture the between-sensor dependence. Then a composite anomaly metric is established with the learned dependence structure explicitly. The experiments on four real-world datasets show our superiority in detection accuracy, anomaly diagnosis, and model interpretation.

PDF Details DOI

YNIMG Journal 2010 Journal Article

Emotion specific modulation of automatic amygdala responses by 5-HTTLPR genotype

Udo Dannlowski
Carsten Konrad
Harald Kugel
Pienie Zwitserlood
Katharina Domschke
Sonja Schöning
Patricia Ohrmann
Jochen Bauer

A functional polymorphism in the serotonin transporter gene (5-HTTLPR) has been reported to modulate amygdala responsiveness to negative environmental cues. However, it remains unclear whether 5-HTTLPR modulates amygdala responses specifically to negative stimuli or rather to emotionally salient stimuli in general. In 44 healthy subjects, amygdala responses to subliminally presented happy and sad facial expressions were assessed by means of fMRI at 3 Tesla. All subjects were genotyped for 5-HTTLPR and the recently discovered 5-HTT rs25531. We observed a robust emotion by genotype group interaction in the right amygdala. Risk allele carriers (S or LG) showed similar amygdala responses to happy faces compared to homozygous LALA carriers but increased amygdala responses to sad faces. The right amygdala was the only anatomical region across the whole brain demonstrating this interaction at a reasonable threshold. It appears that whereas 5-HTT gene variation modulates automatic amygdala responsiveness to sad faces, no such association was found for happy faces. We conclude that 5-HTTLPR genotype predominantly impacts the central processing predominantly of negative environmental cues but not of emotionally salient stimuli in general.

Details DOI