Publications (59)
Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning
Ke Yu, Chao Dong, Liang Lin +1
We investigate a novel approach for image restoration by reinforcement learning. Unlike existing studies that mostly train a single large network for a specialized task, we prepare…
HELIX: Hybrid Encoding with Learnable Identity and Cross-dimensional Synthesis for Time Series Imputation
Fengming Zhang, Wenjie Du, Huan Zhang +2
Time series imputation benefits from leveraging cross-feature correlations, yet existing attention-based methods re-discover feature relationships at each layer, lacking persistent…
Multiscale Cross-Modal Mapping of Molecular, Pathologic, and Radiologic Phenotypes in Lipid-Deficient Clear Cell Renal CellCarcinoma
Ying Cui, Dongzhe Zheng, Ke Yu +8
Clear cell renal cell carcinoma (ccRCC) exhibits extensive intratumoral heterogeneity on multiple biological scales, contributing to variable clinical outcomes and limiting the eff…
DrasCLR: A Self-supervised Framework of Learning Disease-related and Anatomy-specific Representation for 3D Medical Images
Ke Yu, Li Sun, Junxiang Chen +3
Large-scale volumetric medical images with annotation are rare, costly, and time prohibitive to acquire. Self-supervised learning (SSL) offers a promising pre-training and feature…
IRIS Si IV Line Profiles at Flare Ribbons as Indications of Chromospheric Condensation
Ke Yu, Y. Li, M. D. Ding +3
We present temporal variations of the Si IV line profiles at the flare ribbons in three solar flares observed by the Interface Region Imaging Spectrograph (IRIS). In the M1.1 flare…
Beyond Distribution Shift: Spurious Features Through the Lens of Training Dynamics
Nihal Murali, Aahlad Puli, Ke Yu +2
Deep Neural Networks (DNNs) are prone to learning spurious features that correlate with the label during training but are irrelevant to the learning problem. This hurts model gener…
Deep Network Interpolation for Continuous Imagery Effect Transition
Xintao Wang, Ke Yu, Chao Dong +2
Deep convolutional neural network has demonstrated its capability of learning a deterministic mapping for the desired imagery effect. However, the large variety of user flavors mot…
Path-Restore: Learning Network Path Selection for Image Restoration
Ke Yu, Xintao Wang, Chao Dong +2
Very deep Convolutional Neural Networks (CNNs) have greatly improved the performance on various image restoration tasks. However, this comes at a price of increasing computational…
Can contrastive learning avoid shortcut solutions?
Joshua Robinson, Li Sun, Ke Yu +3
The generalization of representations learned via contrastive learning depends crucially on what features of the data are extracted. However, we observe that the contrastive loss d…
BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond
Kelvin C. K. Chan, Xintao Wang, Ke Yu +2
Video super-resolution (VSR) approaches tend to have more components than the image counterparts as they need to exploit the additional temporal dimension. Complex designs are not…
Training-Free Dual Hyperbolic Adapters for Better Cross-Modal Reasoning
Yi Zhang, Chun-Wun Cheng, Junyi He +5
Recent research in Vision-Language Models (VLMs) has significantly advanced our capabilities in cross-modal reasoning. However, existing methods suffer from performance degradation…
Temperature dependence of quasi-localized phonons-mediated non-Markovianity dynamics of SiV^- centers in diamond
Wanggui Ye, Debao Zhang, Xuguang Cao +6
Here we investigate the temperature-dependent non-Markovian dynamics of the SiV^- center in diamond, focusing on the roles of low- and high-frequency quasi-localized phonon modes.…
Concept-Guided Prompt Learning for Generalization in Vision-Language Models
Yi Zhang, Ce Zhang, Ke Yu +2
Contrastive Language-Image Pretraining (CLIP) model has exhibited remarkable efficacy in establishing cross-modal connections between texts and images, yielding impressive performa…
Considering Spatial Structure of the Road Network in Pavement Deterioration Modeling
Lu Gao, Ke Yu, Pan Lu
Pavement deterioration modeling is important in providing information regarding the future state of the road network and in determining the needs of preventive maintenance or rehab…
Learning and Current Prediction of PMSM Drive via Differential Neural Networks
Wenjie Mei, Xiaorui Wang, Yanrong Lu +2
Learning models for dynamical systems in continuous time is significant for understanding complex phenomena and making accurate predictions. This study presents a novel approach ut…
Improving Text Matching in E-Commerce Search with A Rationalizable, Intervenable and Fast Entity-Based Relevance Model
Jiong Cai, Yong Jiang, Yue Zhang +10
Discovering the intended items of user queries from a massive repository of items is one of the main goals of an e-commerce search system. Relevance prediction is essential to the…
Asset Allocation and Risk Assessment with Gross Exposure Constraints for Vast Portfolios
Jianqing Fan, Jingjin Zhang, Ke Yu
Markowitz (1952, 1959) laid down the ground-breaking work on the mean-variance analysis. Under his framework, the theoretical optimal allocation vector can be very different from t…
On the origin of a broad QFP wave train: unwinding jet as the driver
Xinping Zhou, Zehao Tang, Zhining Qu +5
Large-scale extreme-ultraviolet (EUV) waves commonly exhibit as single wavefront and are believed to be caused by coronal mass ejections (CMEs). Utilizing high spatiotemporal resol…
SRLR: Symbolic Regression based Logic Recovery to Counter Programmable Logic Controller Attacks
Hao Zhou, Suman Sourav, Binbin Chen +1
Programmable Logic Controllers (PLCs) are critical components in Industrial Control Systems (ICSs). Their potential exposure to external world makes them susceptible to cyber-attac…
Dynamics and decay of a spherical region of turbulence in free space
Ke Yu, Tim Colonius, D. I. Pullin +1
We perform direct numerical simulation (DNS) and large eddy simulation (LES) of an initially spherical region of turbulence evolving in free space. The computations are performed w…
Hierarchical Amortized Training for Memory-efficient High Resolution 3D GAN
Li Sun, Junxiang Chen, Yanwu Xu +3
Generative Adversarial Networks (GAN) have many potential medical imaging applications, including data augmentation, domain adaptation, and model explanation. Due to the limited me…
Conceptual Codebook Learning for Vision-Language Models
Yi Zhang, Ke Yu, Siqi Wu +1
In this paper, we propose Conceptual Codebook Learning (CoCoLe), a novel fine-tuning method for vision-language models (VLMs) to address the challenge of improving the generalizati…
Boosting the interpretability of clinical risk scores with intervention predictions
Eric Loreaux, Ke Yu, Jonas Kemp +8
Machine learning systems show significant promise for forecasting patient adverse events via risk scores. However, these risk scores implicitly encode assumptions about future inte…
Learning to Adapt CLIP for Few-Shot Monocular Depth Estimation
Xueting Hu, Ce Zhang, Yi Zhang +3
Pre-trained Vision-Language Models (VLMs), such as CLIP, have shown enhanced performance across a range of tasks that involve the integration of visual and linguistic modalities. W…
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
Xintao Wang, Kelvin C. K. Chan, Ke Yu +2
Video restoration tasks, including super-resolution, deblurring, etc, are drawing increasing attention in the computer vision community. A challenging benchmark named REDS is relea…
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
Xintao Wang, Ke Yu, Shixiang Wu +6
The Super-Resolution Generative Adversarial Network (SRGAN) is a seminal work that is capable of generating realistic textures during single image super-resolution. However, the ha…
Anatomy-Guided Weakly-Supervised Abnormality Localization in Chest X-rays
Ke Yu, Shantanu Ghosh, Zhexiong Liu +2
Creating a large-scale dataset of abnormality annotation on medical images is a labor-intensive and costly task. Leveraging weak supervision from readily available data such as rad…
Imaging and Spectroscopic Observations of the Dynamic Processes in Limb Solar Flares
Ke Yu, Y. Li, Jie Hong +2
We investigate various dynamic processes including magnetic reconnection, chromospheric evaporation, and coronal rain draining in two limb solar flares through imaging and spectros…
Distilling BlackBox to Interpretable models for Efficient Transfer Learning
Shantanu Ghosh, Ke Yu, Kayhan Batmanghelich
Building generalizable AI models is one of the primary challenges in the healthcare domain. While radiologists rely on generalizable descriptive rules of abnormality, Neural Networ…
Rethinking Text-based Protein Understanding: Retrieval or LLM?
Juntong Wu, Zijing Liu, He Cao +6
In recent years, protein-text models have gained significant attention for their potential in protein generation and understanding. Current approaches focus on integrating protein-…
Tackling Shortcut Learning in Deep Neural Networks: An Iterative Approach with Interpretable Models
Shantanu Ghosh, Ke Yu, Forough Arabshahi +1
We use concept-based interpretable models to mitigate shortcut learning. Existing methods lack interpretability. Beginning with a Blackbox, we iteratively carve out a mixture of in…
Two-sided-loop jet originates from the filament internal reconnection
Yunxue Huang, Jialin Li, Zhining Qu +4
Magnetic reconnection driving two-sided-loop jet is typically associated with interactions between an emerging bipole and the overlying horizontal magnetic field, or between filame…
ACPs: Agent Collaboration Protocols for the Internet of Agents
Jun Liu, Ke Yu, Keliang Chen +5
With the rapid advancement of artificial intelligence, the proliferation of autonomous agents has introduced new challenges in interoperability, scalability, and coordination. The…
Deep Convolution Networks for Compression Artifacts Reduction
Ke Yu, Chao Dong, Chen Change Loy +1
Lossy compression introduces complex compression artifacts, particularly blocking artifacts, ringing effects and blurring. Existing algorithms either focus on removing blocking art…
Broad and Bi-directional narrow quasi-periodic fast-propagating wave trains associated with a filament-driven halo CME on 2023 April 21
Xinping Zhou, Yuandeng Shen, Yihua Yan +8
This paper presents three distinct wave trains that occurred on 2023 April 21: a broad quasi-periodic fast-propagating (QFP) wave train and a bi-directional narrow QFP wave train.…
Two-stage primary acceleration in filament initial eruption under a fan-spine magnetic configuration
Haitang Li, Ke Yu, Chang Zhou +6
Understaning the filament rising process is crucial for unveiling the triggering mechanisms of the coronal mass ejections and forecasting the space weather. In this paper, we prese…
Hyperbolic Molecular Representation Learning for Drug Repositioning
Ke Yu, Shyam Visweswaran, Kayhan Batmanghelich
Learning accurate drug representations is essential for task such as computational drug repositioning. A drug hierarchy is a valuable source that encodes knowledge of relations amo…
A fast multi-resolution lattice Green's function method for elliptic difference equations
Benedikt Dorschner, Ke Yu, Gianmarco Mengaldo +1
We propose a mesh refinement technique for solving elliptic difference equations on unbounded domains based on the fast lattice Green's function (FLGF) method. The FLGF method expl…
Two-Step Active Learning for Instance Segmentation with Uncertainty and Diversity Sampling
Ke Yu, Stephen Albro, Giulia DeSalvo +5
Training high-quality instance segmentation models requires an abundance of labeled images with instance masks and classifications, which is often expensive to procure. Active lear…
Dividing and Conquering a BlackBox to a Mixture of Interpretable Models: Route, Interpret, Repeat
Shantanu Ghosh, Ke Yu, Forough Arabshahi +1
ML model design either starts with an interpretable model or a Blackbox and explains it post hoc. Blackbox models are flexible but difficult to explain, while interpretable models…
Context Matters: Graph-based Self-supervised Representation Learning for Medical Images
Li Sun, Ke Yu, Kayhan Batmanghelich
Supervised learning method requires a large volume of annotated datasets. Collecting such datasets is time-consuming and expensive. Until now, very few annotated COVID-19 imaging d…
Direct Observation of Whispering Gallery Mode Polaritons and Their Dispersion in a ZnO Tapered Microcavity
Liaoxin Sun, Zhanghai Chen, Qijun Ren +6
We report direct observation of the strong exciton-photon coupling in ZnO tapered whispering gallery (WG) microcavity at room temperature. By scanning excitations along the tapered…
NODE-Adapter: Neural Ordinary Differential Equations for Better Vision-Language Reasoning
Yi Zhang, Chun-Wun Cheng, Ke Yu +3
In this paper, we consider the problem of prototype-based vision-language reasoning problem. We observe that existing methods encounter three major challenges: 1) escalating resour…
SHAP Distance: An Explainability-Aware Metric for Evaluating the Semantic Fidelity of Synthetic Tabular Data
Ke Yu, Shigeru Ishikura, Yukari Usukura +2
Synthetic tabular data, which are widely used in domains such as healthcare, enterprise operations, and customer analytics, are increasingly evaluated to ensure that they preserve…
GraphFedMIG: Tackling Class Imbalance in Federated Graph Learning via Mutual Information-Guided Generation
Xinrui Li, Qilin Fan, Tianfu Wang +3
Federated graph learning (FGL) enables multiple clients to collaboratively train powerful graph neural networks without sharing their private, decentralized graph data. Inherited f…
ReconfigISP: Reconfigurable Camera Image Processing Pipeline
Ke Yu, Zexian Li, Yue Peng +2
Image Signal Processor (ISP) is a crucial component in digital cameras that transforms sensor signals into images for us to perceive and understand. Existing ISP designs always ado…
Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform
Xintao Wang, Ke Yu, Chao Dong +1
Despite that convolutional neural networks (CNN) have recently demonstrated high-quality reconstruction for single-image super-resolution (SR), recovering natural and realistic tex…
Semi-Supervised Hierarchical Drug Embedding in Hyperbolic Space
Ke Yu, Shyam Visweswaran, Kayhan Batmanghelich
Learning accurate drug representation is essential for tasks such as computational drug repositioning and prediction of drug side-effects. A drug hierarchy is a valuable source tha…
Lattice Discrete Particle Model (LDPM): Comparison of Various Time Integration Solvers and Implementations
Erol Lale, Jan Eliáš, Ke Yu +17
This article presents a comparison of various implementations of the Lattice Discrete Particle Model (LDPM) for the numerical simulation of concrete and other heterogeneous quasibr…
Understanding Deformable Alignment in Video Super-Resolution
Kelvin C. K. Chan, Xintao Wang, Ke Yu +2
Deformable convolution, originally proposed for the adaptation to geometric variations of objects, has recently shown compelling performance in aligning multiple frames and is incr…
A Cautionary Tale of Self-Supervised Learning for Imaging Biomarkers: Alzheimer's Disease Case Study
Maxwell Reynolds, Chaitanya Srinivasan, Vijay Cherupally +6
Discovery of sensitive and biologically grounded biomarkers is essential for early detection and monitoring of Alzheimer's disease (AD). Structural MRI is widely available but typi…
3D fast-mode Wave Propagation from Corona to Chromosphere: Triggering Mechanism for 3D Oscillations of filaments
Yunxue Huang, Qin Feng, Yuhu Miao +5
Moreton waves are widely regarded as the chromospheric counterpart of extreme ultraviolet (EUV) waves propagating in the corona. However, direct observational evidence confirming t…
Spectroscopic Observations of High-speed Downflows in a C1.7 Solar Flare
Yi-An Zhou, Y. Li, M. D. Ding +2
In this paper, we analyze the high-resolution UV spectra for a C1.7 solar flare (SOL2017-09-09T06:51) observed by the \textit{Interface Region Imaging Spectrograph} (\textit{IRIS})…
Pavement Missing Condition Data Imputation through Collective Learning-Based Graph Neural Networks
Ke Yu, Lu Gao
Pavement condition data is important in providing information regarding the current state of the road network and in determining the needs of maintenance and rehabilitation treatme…
Context-aware Self-supervised Learning for Medical Images Using Graph Neural Network
Li Sun, Ke Yu, Kayhan Batmanghelich
Although self-supervised learning enables us to bootstrap the training by exploiting unlabeled data, the generic self-supervised methods for natural images do not sufficiently inco…
ICODE: Modeling Dynamical Systems with Extrinsic Input Information
Zhaoyi Li, Wenjie Mei, Ke Yu +2
Learning models of dynamical systems with external inputs, which may be, for example, nonsmooth or piecewise, is crucial for studying complex phenomena and predicting future state…
Improving On-policy Learning with Statistical Reward Accumulation
Yubin Deng, Ke Yu, Dahua Lin +2
Deep reinforcement learning has obtained significant breakthroughs in recent years. Most methods in deep-RL achieve good results via the maximization of the reward signal provided…
Multi-resolution lattice Green's function method for incompressible flows
Ke Yu, Benedikt Dorschner, Tim Colonius
We propose a multi-resolution strategy that is compatible with the lattice Green's function (LGF) technique for solving viscous, incompressible flows on unbounded domains. The LGF…
Vast Volatility Matrix Estimation using High Frequency Data for Portfolio Selection
Jianqing Fan, Yingying Li, Ke Yu
Portfolio allocation with gross-exposure constraint is an effective method to increase the efficiency and stability of selected portfolios among a vast pool of assets, as demonstra…