Sai Wu
YOU?
Author Swipe
View article: RLOMM: An Efficient and Robust Online Map Matching Framework with Reinforcement Learning
RLOMM: An Efficient and Robust Online Map Matching Framework with Reinforcement Learning Open
Online map matching is a fundamental problem in location-based services, aiming to incrementally match trajectory data step-by-step onto a road network. However, existing methods fail to meet the needs for efficiency, robustness, and accur…
View article: A Quantum Framework for Combinatorial Optimization Problem over Graphs
A Quantum Framework for Combinatorial Optimization Problem over Graphs Open
Combinatorial optimization problems over graphs, such as the traveling salesman problem, longest path problem, and maximum independent set problem, are well-known for being computationally costly, some even NP-hard problems. In this paper,…
View article: AIGT: AI Generative Table Based on Prompt
AIGT: AI Generative Table Based on Prompt Open
Tabular data, which accounts for over 80% of enterprise data assets, is vital in various fields. With growing concerns about privacy protection and data-sharing restrictions, generating high-quality synthetic tabular data has become essent…
View article: Ultra‐Broadband and Reconfigurable Liquid‐Based Microwave Metasurface Absorber
Ultra‐Broadband and Reconfigurable Liquid‐Based Microwave Metasurface Absorber Open
Liquids, with their advantages of fluidity, ease of shaping, and tunability, have exhibited promising potential in the creation of reconfigurable metamaterials (MMs). Water, being the ubiquitous liquid and cost‐effective resource on Earth,…
View article: Single-cell analysis identifies PLK1 as a driver of immunosuppressive tumor microenvironment in LUAD
Single-cell analysis identifies PLK1 as a driver of immunosuppressive tumor microenvironment in LUAD Open
PLK1 (Polo-like kinase 1) plays a critical role in the progression of lung adenocarcinoma (LUAD). Recent studies have unveiled that targeting PLK1 improves the efficacy of immunotherapy, highlighting its important role in the regulation of…
View article: Revisiting CNNs for Trajectory Similarity Learning
Revisiting CNNs for Trajectory Similarity Learning Open
Similarity search is a fundamental but expensive operator in querying trajectory data, due to its quadratic complexity of distance computation. To mitigate the computational burden for long trajectories, neural networks have been widely em…
View article: Quantum Computing for Databases: Overview and Challenges
Quantum Computing for Databases: Overview and Challenges Open
In the decades, the general field of quantum computing has experienced remarkable progress since its inception. A plethora of researchers not only proposed quantum algorithms showing the power of quantum computing but also constructed the …
View article: Against Jamming Attack in Wireless Communication Networks: A Reinforcement Learning Approach
Against Jamming Attack in Wireless Communication Networks: A Reinforcement Learning Approach Open
When wireless communication networks encounter jamming attacks, they experience spectrum resource occupation and data communication failures. In order to address this issue, an anti-jamming algorithm based on distributed multi-agent reinfo…
View article: Sampling-Resilient Multi-Object Tracking
Sampling-Resilient Multi-Object Tracking Open
Multi-Object Tracking (MOT) is a cornerstone operator for video surveillance applications. To enable real-time processing of large-scale live video streams, we study an interesting scenario called down-sampled MOT, which performs object tr…
View article: FL-GUARD: A Holistic Framework for Run-Time Detection and Recovery of Negative Federated Learning
FL-GUARD: A Holistic Framework for Run-Time Detection and Recovery of Negative Federated Learning Open
Federated learning (FL) is a promising approach for learning a model from data distributed on massive clients without exposing data privacy. It works effectively in the ideal federation where clients share homogeneous data distribution and…
View article: MAGPIE: accurate pathogenic prediction for multiple variant types using machine learning approach
MAGPIE: accurate pathogenic prediction for multiple variant types using machine learning approach Open
Identifying pathogenic variants from the vast majority of nucleotide variation remains a challenge. We present a method named Multimodal Annotation Generated Pathogenic Impact Evaluator (MAGPIE) that predicts the pathogenicity of multi-typ…
View article: Phosphorylation of AHR by PLK1 promotes metastasis of LUAD via DIO2-TH signaling
Phosphorylation of AHR by PLK1 promotes metastasis of LUAD via DIO2-TH signaling Open
Metastasis of lung adenocarcinoma (LUAD) is a major cause of death in patients. Aryl hydrocarbon receptor (AHR), an important transcription factor, is involved in the initiation and progression of lung cancer. Polo-like kinase 1 (PLK1), a …
View article: ModelGiF: Gradient Fields for Model Functional Distance
ModelGiF: Gradient Fields for Model Functional Distance Open
The last decade has witnessed the success of deep learning and the surge of publicly released trained models, which necessitates the quantification of the model functional distance for various purposes. However, quantifying the model funct…
View article: Single-cell analysis characterizes PLK1 as a catalyst of an immunosuppressive tumor microenvironment in LUAD
Single-cell analysis characterizes PLK1 as a catalyst of an immunosuppressive tumor microenvironment in LUAD Open
PLK1 (Polo-like kinase 1) plays a critical role in the progression of lung adenocarcinoma (LUAD). Recent studies have unveiled that targeting PLK1 improves the efficacy of immunotherapy, highlighting its important role in the regulation of…
View article: Phosphorylation of AHR by PLK1 promotes metastasis of LUAD via DIO2-TH signaling
Phosphorylation of AHR by PLK1 promotes metastasis of LUAD via DIO2-TH signaling Open
Metastasis of Lung adenocarcinoma (LUAD) is a major cause of death in patients. Aryl hydrocarbon receptor (AHR) is an important transcription factor involved in the initiation and progression of lung cancer. Polo-like kinase 1 (PLK1), a se…
View article: Towards Cross-Table Masked Pretraining for Web Data Mining
Towards Cross-Table Masked Pretraining for Web Data Mining Open
Tabular data pervades the landscape of the World Wide Web, playing a foundational role in the digital architecture that underpins online information. Given the recent influence of large-scale pretrained models like ChatGPT and SAM across v…
View article: Assessing Hidden Risks of LLMs: An Empirical Study on Robustness, Consistency, and Credibility
Assessing Hidden Risks of LLMs: An Empirical Study on Robustness, Consistency, and Credibility Open
The recent popularity of large language models (LLMs) has brought a significant impact to boundless fields, particularly through their open-ended ecosystem such as the APIs, open-sourced models, and plugins. However, with their widespread …
View article: Controllable Textual Inversion for Personalized Text-to-Image Generation
Controllable Textual Inversion for Personalized Text-to-Image Generation Open
The recent large-scale generative modeling has attained unprecedented performance especially in producing high-fidelity images driven by text prompts. Text inversion (TI), alongside the text-to-image model backbones, is proposed as an effe…
View article: Byzantine-Robust Learning on Heterogeneous Data via Gradient Splitting
Byzantine-Robust Learning on Heterogeneous Data via Gradient Splitting Open
Federated learning has exhibited vulnerabilities to Byzantine attacks, where the Byzantine attackers can send arbitrary gradients to a central server to destroy the convergence and performance of the global model. A wealth of robust AGgreg…
View article: A Survey on Mapping Semi-Structured Data and Graph Data to Relational Data
A Survey on Mapping Semi-Structured Data and Graph Data to Relational Data Open
The data produced by various services should be stored and managed in an appropriate format for gaining valuable knowledge conveniently. This leads to the emergence of various data models, including relational, semi-structured, and graph m…
View article: Comparison Knowledge Translation for Generalizable Image Classification
Comparison Knowledge Translation for Generalizable Image Classification Open
Deep learning has recently achieved remarkable performance in image classification tasks, which depends heavily on massive annotation. However, the classification mechanism of existing deep learning models seems to contrast to humans' reco…
View article: Comparison Knowledge Translation for Generalizable Image Classification
Comparison Knowledge Translation for Generalizable Image Classification Open
Deep learning has recently achieved remarkable performance in image classification tasks, which depends heavily on massive annotation. However, the classification mechanism of existing deep learning models seems to contrast to humans' reco…
View article: Dynamic Index Construction with Deep Reinforcement Learning
Dynamic Index Construction with Deep Reinforcement Learning Open
Thanks to the rapid advances in artificial intelligence, a brand new venue for database performance optimization is through deep neural networks and the reinforcement learning paradigm. Alongside the long literature in this regime, an icon…
View article: Towards Unifying the Label Space for Aspect- and Sentence-based Sentiment Analysis
Towards Unifying the Label Space for Aspect- and Sentence-based Sentiment Analysis Open
The aspect-based sentiment analysis (ABSA) is a fine-grained task that aims to determine the sentiment polarity towards targeted aspect terms occurring in the sentence. The development of the ABSA task is very much hindered by the lack of …
View article: Model Doctor: A Simple Gradient Aggregation Strategy for Diagnosing and Treating CNN Classifiers
Model Doctor: A Simple Gradient Aggregation Strategy for Diagnosing and Treating CNN Classifiers Open
Recently, Convolutional Neural Network (CNN) has achieved excellent performance in the classification task. It is widely known that CNN is deemed as a 'black-box', which is hard for understanding the prediction mechanism and debugging the …
View article: Joining datasets via data augmentation in the label space for neural networks
Joining datasets via data augmentation in the label space for neural networks Open
Most, if not all, modern deep learning systems restrict themselves to a single dataset for neural network training and inference. In this article, we are interested in systematic ways to join datasets that are made of similar purposes. Unl…