Wei Emma Zhang
YOU?
Author Swipe
View article: Adaptive Spline Networks in the Kolmogorov-Arnold Framework: Knot Analysis and Stability Enhancement
Adaptive Spline Networks in the Kolmogorov-Arnold Framework: Knot Analysis and Stability Enhancement Open
View article: MMiC: Mitigating Modality Incompleteness in Clustered Federated Learning
MMiC: Mitigating Modality Incompleteness in Clustered Federated Learning Open
In the era of big data, data mining has become indispensable for uncovering hidden patterns and insights from vast and complex datasets. The integration of multimodal data sources further enhances its potential. Multimodal Federated Learni…
View article: Calibrating on Medical Segmentation Model through Signed Distance
Calibrating on Medical Segmentation Model through Signed Distance Open
View article: Calibrating on Kolmogorov-Arnold Network
Calibrating on Kolmogorov-Arnold Network Open
View article: FedDPGu: Adaptive Prompt-tuning with Built-in Unlearning for Federated Learning
FedDPGu: Adaptive Prompt-tuning with Built-in Unlearning for Federated Learning Open
Pre-trained Language Models (PLMs) have demonstrated impressive performance in various NLP tasks. However, traditional fine-tuning methods for leveraging PLMs for downstream tasks entail significant computational overhead. Prompt-tuning ha…
View article: Lifting Manifolds to Mitigate Pseudo-Alignment in LLM4TS
Lifting Manifolds to Mitigate Pseudo-Alignment in LLM4TS Open
Pseudo-Alignment is a pervasive challenge in many large language models for time series (LLM4TS) models, often causing them to underperform compared to linear models or randomly initialised backbones. However, there is limited discussion i…
View article: GENETIC DIVERSITY ANALYSIS OF NORTHERN <italic>RHYNCHOCYPRIS LAGOWSKII</italic> POPULATIONS BASED ON GENOTYPING BY SEQUENCING TECHNOLOGY
GENETIC DIVERSITY ANALYSIS OF NORTHERN <italic>RHYNCHOCYPRIS LAGOWSKII</italic> POPULATIONS BASED ON GENOTYPING BY SEQUENCING TECHNOLOGY Open
View article: The rational decommissioning of solar photovoltaic (PV) technologies: evidence from Australia
The rational decommissioning of solar photovoltaic (PV) technologies: evidence from Australia Open
View article: FediLoRA: Heterogeneous LoRA for Federated Multimodal Fine-tuning under Missing Modalities
FediLoRA: Heterogeneous LoRA for Federated Multimodal Fine-tuning under Missing Modalities Open
Foundation models have demonstrated remarkable performance across a wide range of tasks, yet their large parameter sizes pose challenges for practical deployment, especially in decentralized environments. Parameter-efficient fine-tuning (P…
View article: Calibration Attention: Instance-wise Temperature Scaling for Vision Transformers
Calibration Attention: Instance-wise Temperature Scaling for Vision Transformers Open
Probability calibration is critical when Vision Transformers are deployed in risk-sensitive applications. The standard fix, post-hoc temperature scaling, uses a single global scalar and requires a held-out validation set. We introduce Cali…
View article: Understanding Why Large Language Models Can Be Ineffective in Time Series Analysis: The Impact of Modality Alignment
Understanding Why Large Language Models Can Be Ineffective in Time Series Analysis: The Impact of Modality Alignment Open
View article: SCOPE: Compress Mathematical Reasoning Steps for Efficient Automated Process Annotation
SCOPE: Compress Mathematical Reasoning Steps for Efficient Automated Process Annotation Open
Process Reward Models (PRMs) have demonstrated promising results in mathematical reasoning, but existing process annotation approaches, whether through human annotations or Monte Carlo simulations, remain computationally expensive. In this…
View article: A survey of deep-learning-based radiology report generation using multimodal inputs
A survey of deep-learning-based radiology report generation using multimodal inputs Open
Automatic radiology report generation can alleviate the workload for physicians and minimize regional disparities in medical resources, therefore becoming an important topic in the medical image analysis field. It is a challenging task, as…
View article: Electricity Cost Minimization for Multi-Workflow Allocation in Geo-Distributed Data Centers
Electricity Cost Minimization for Multi-Workflow Allocation in Geo-Distributed Data Centers Open
Worldwide, Geo-distributed Data Centers (GDCs) provide computing and storage services for massive workflow applications, resulting in high electricity costs that vary depending on geographical locations and time. How to reduce electricity …
View article: How do environmental and operational factors impact particulate matter dynamics in building construction? - Insights from real-time sensing
How do environmental and operational factors impact particulate matter dynamics in building construction? - Insights from real-time sensing Open
Construction activities are major contributors to particulate matter (PM) pollution, posing significant risks to workers and nearby populations. However, PM mitigation strategies in the complex nature of construction environments remain un…
View article: PostHoc FREE Calibrating on Kolmogorov Arnold Networks
PostHoc FREE Calibrating on Kolmogorov Arnold Networks Open
Kolmogorov Arnold Networks (KANs) are neural architectures inspired by the Kolmogorov Arnold representation theorem that leverage B Spline parameterizations for flexible, locally adaptive function approximation. Although KANs can capture c…
View article: TrojanTime: Backdoor Attacks on Time Series Classification
TrojanTime: Backdoor Attacks on Time Series Classification Open
Time Series Classification (TSC) is highly vulnerable to backdoor attacks, posing significant security threats. Existing methods primarily focus on data poisoning during the training phase, designing sophisticated triggers to improve steal…
View article: Free-Knots Kolmogorov-Arnold Network: On the Analysis of Spline Knots and Advancing Stability
Free-Knots Kolmogorov-Arnold Network: On the Analysis of Spline Knots and Advancing Stability Open
Kolmogorov-Arnold Neural Networks (KANs) have gained significant attention in the machine learning community. However, their implementation often suffers from poor training stability and heavy trainable parameter. Furthermore, there is lim…
View article: FedDPG: An Adaptive Yet Efficient Prompt-Tuning Approach in Federated Learning Settings
FedDPG: An Adaptive Yet Efficient Prompt-Tuning Approach in Federated Learning Settings Open
View article: Evolution of Structural Properties and its Determinants of Global Waste Trade Network Based on Temporal Exponential Random Graph Models
Evolution of Structural Properties and its Determinants of Global Waste Trade Network Based on Temporal Exponential Random Graph Models Open
View article: The More, The Better? A Critical Study of Multimodal Context in Radiology Report Summarization
The More, The Better? A Critical Study of Multimodal Context in Radiology Report Summarization Open
View article: Can VLMs Actually See and Read? A Survey on Modality Collapse in Vision-Language Models
Can VLMs Actually See and Read? A Survey on Modality Collapse in Vision-Language Models Open
View article: Challenges and Prospects for Deepening Coordinated Development of the Beijing-Tianjin-Hebei Region
Challenges and Prospects for Deepening Coordinated Development of the Beijing-Tianjin-Hebei Region Open
View article: SCOPE: Compress Mathematical Reasoning Steps for Efficient Automated Process Annotation
SCOPE: Compress Mathematical Reasoning Steps for Efficient Automated Process Annotation Open
View article: Fine-Tuning Encoder-Decoder Models with Contrastive Learning for In-Context Distractor Generation
Fine-Tuning Encoder-Decoder Models with Contrastive Learning for In-Context Distractor Generation Open
View article: Deep learning meets bibliometrics: A survey of citation function classification
Deep learning meets bibliometrics: A survey of citation function classification Open
View article: Cognitive Digital Twins for the Microgrid: A Real-World Study for Intelligent Energy Management and Optimization
Cognitive Digital Twins for the Microgrid: A Real-World Study for Intelligent Energy Management and Optimization Open
Digital twin technology is a promising solution for achieving optimized microgrid control with enhanced efficiency, reliability, and sustainability. In this paper, we focus on a real-world microgrid in Singapore and develop a cognitive dig…
View article: Calibrating Deep Neural Network using Euclidean Distance
Calibrating Deep Neural Network using Euclidean Distance Open
Uncertainty is a fundamental aspect of real-world scenarios, where perfect information is rarely available. Humans naturally develop complex internal models to navigate incomplete data and effectively respond to unforeseen or partially obs…
View article: Devil in the Tail: A Multi-Modal Framework for Drug-Drug Interaction Prediction in Long Tail Distinction
Devil in the Tail: A Multi-Modal Framework for Drug-Drug Interaction Prediction in Long Tail Distinction Open
View article: Boosting Certificate Robustness for Time Series Classification with Efficient Self-Ensemble
Boosting Certificate Robustness for Time Series Classification with Efficient Self-Ensemble Open
Recently, the issue of adversarial robustness in the time series domain has garnered significant attention. However, the available defense mechanisms remain limited, with adversarial training being the predominant approach, though it does …