Shi Ye
YOU?
Author Swipe
View article: Diffusion Bridge or Flow Matching? A Unifying Framework and Comparative Analysis
Diffusion Bridge or Flow Matching? A Unifying Framework and Comparative Analysis Open
Diffusion Bridge and Flow Matching have both demonstrated compelling empirical performance in transformation between arbitrary distributions. However, there remains confusion about which approach is generally preferable, and the substantia…
View article: Efficiency evaluation of fuel retention diagnostic in first wall by LID-QMS: Based on LIBS
Efficiency evaluation of fuel retention diagnostic in first wall by LID-QMS: Based on LIBS Open
View article: Smart phosphor with neuromorphic behaviors enabling full-photoluminescent Write and Read for all-optical physical reservoir computing
Smart phosphor with neuromorphic behaviors enabling full-photoluminescent Write and Read for all-optical physical reservoir computing Open
The unprecedented growth in information across diverse media drives an urgent need for multifunctional materials and devices beyond conventional electrical paradigms. This work explores all-optical information processing based on photolumi…
View article: FreqPolicy: Frequency Autoregressive Visuomotor Policy with Continuous Tokens
FreqPolicy: Frequency Autoregressive Visuomotor Policy with Continuous Tokens Open
Learning effective visuomotor policies for robotic manipulation is challenging, as it requires generating precise actions while maintaining computational efficiency. Existing methods remain unsatisfactory due to inherent limitations in the…
View article: Exploring the Boundary of Diffusion-based Methods for Solving Constrained Optimization
Exploring the Boundary of Diffusion-based Methods for Solving Constrained Optimization Open
Diffusion models have achieved remarkable success in generative tasks such as image and video synthesis, and in control domains like robotics, owing to their strong generalization capabilities and proficiency in fitting complex multimodal …
View article: UniDB: A Unified Diffusion Bridge Framework via Stochastic Optimal Control
UniDB: A Unified Diffusion Bridge Framework via Stochastic Optimal Control Open
Recent advances in diffusion bridge models leverage Doob's $h$-transform to establish fixed endpoints between distributions, demonstrating promising results in image translation and restoration tasks. However, these approaches frequently p…
View article: Boosting the Photocatalytic Performance for Oxidative Coupling of Amines to Imines by Fabricating the Zn0.1cd0.9s/In2o3 Heterojunction
Boosting the Photocatalytic Performance for Oxidative Coupling of Amines to Imines by Fabricating the Zn0.1cd0.9s/In2o3 Heterojunction Open
View article: Boosting the Photocatalytic Performance for Oxidative Coupling of Amines to Imines by Fabricating the Zn0.1cd0.9s/In2o3 Heterojunction
Boosting the Photocatalytic Performance for Oxidative Coupling of Amines to Imines by Fabricating the Zn0.1cd0.9s/In2o3 Heterojunction Open
View article: Numerical simulation of GTAW for ZW61 magnesium alloy thin plates: Coupling the finite element method with the cellular automata method
Numerical simulation of GTAW for ZW61 magnesium alloy thin plates: Coupling the finite element method with the cellular automata method Open
ZW61 magnesium alloy has a wide range of application prospects as a lightweight green engineering material. In this paper, the temperature field and microstructure of gas tungsten arc welding (GTAW) for ZW61 magnesium alloy are simulated b…
View article: Harmonizing Generalization and Personalization in Federated Prompt Learning
Harmonizing Generalization and Personalization in Federated Prompt Learning Open
Federated Prompt Learning (FPL) incorporates large pre-trained Vision-Language models (VLM) into federated learning through prompt tuning. The transferable representations and remarkable generalization capacity of VLM make them highly comp…
View article: THOR: Text to Human-Object Interaction Diffusion via Relation Intervention
THOR: Text to Human-Object Interaction Diffusion via Relation Intervention Open
This paper addresses new methodologies to deal with the challenging task of generating dynamic Human-Object Interactions from textual descriptions (Text2HOI). While most existing works assume interactions with limited body parts or static …
View article: Comprehensive Profiling of Acetylcholinesterase Inhibitors from Fried Centipede Using Activity-Oriented Online Preparation Technology
Comprehensive Profiling of Acetylcholinesterase Inhibitors from Fried Centipede Using Activity-Oriented Online Preparation Technology Open
View article: Effects of Different Full-Reference Quality Assessment Metrics in End-to-End Deep Video Coding
Effects of Different Full-Reference Quality Assessment Metrics in End-to-End Deep Video Coding Open
Visual quality assessment is often used as a key performance indicator (KPI) to evaluate the performance of electronic devices. There exists a significant association between visual quality assessment and electronic devices. In this paper,…
View article: IKOL: Inverse Kinematics Optimization Layer for 3D Human Pose and Shape Estimation via Gauss-Newton Differentiation
IKOL: Inverse Kinematics Optimization Layer for 3D Human Pose and Shape Estimation via Gauss-Newton Differentiation Open
This paper presents an inverse kinematic optimization layer (IKOL) for 3D human pose and shape estimation that leverages the strength of both optimization- and regression-based methods within an end-to-end framework. IKOL involves a noncon…
View article: Lifelong Person Re-identification via Knowledge Refreshing and Consolidation
Lifelong Person Re-identification via Knowledge Refreshing and Consolidation Open
Lifelong person re-identification (LReID) is in significant demand for real-world development as a large amount of ReID data is captured from diverse locations over time and cannot be accessed at once inherently. However, a key challenge f…
View article: DreamFace: Progressive Generation of Animatable 3D Faces under Text Guidance
DreamFace: Progressive Generation of Animatable 3D Faces under Text Guidance Open
Emerging Metaverse applications demand accessible, accurate, and easy-to-use tools for 3D digital human creations in order to depict different cultures and societies as if in the physical world. Recent large-scale vision-language advances …
View article: IKOL: Inverse kinematics optimization layer for 3D human pose and shape estimation via Gauss-Newton differentiation
IKOL: Inverse kinematics optimization layer for 3D human pose and shape estimation via Gauss-Newton differentiation Open
This paper presents an inverse kinematic optimization layer (IKOL) for 3D human pose and shape estimation that leverages the strength of both optimization- and regression-based methods within an end-to-end framework. IKOL involves a noncon…
View article: NeuralDome: A Neural Modeling Pipeline on Multi-View Human-Object Interactions
NeuralDome: A Neural Modeling Pipeline on Multi-View Human-Object Interactions Open
Humans constantly interact with objects in daily life tasks. Capturing such processes and subsequently conducting visual inferences from a fixed viewpoint suffers from occlusions, shape and texture ambiguities, motions, etc. To mitigate th…
View article: Lifelong Person Re-Identification via Knowledge Refreshing and Consolidation
Lifelong Person Re-Identification via Knowledge Refreshing and Consolidation Open
Lifelong person re-identification (LReID) is in significant demand for real-world development as a large amount of ReID data is captured from diverse locations over time and cannot be accessed at once inherently. However, a key challenge f…
View article: Knowledge-Aware Federated Active Learning with Non-IID Data
Knowledge-Aware Federated Active Learning with Non-IID Data Open
Federated learning enables multiple decentralized clients to learn collaboratively without sharing the local training data. However, the expensive annotation cost to acquire data labels on local clients remains an obstacle in utilizing loc…
View article: FedTP: Federated Learning by Transformer Personalization
FedTP: Federated Learning by Transformer Personalization Open
Federated learning is an emerging learning paradigm where multiple clients collaboratively train a machine learning model in a privacy-preserving manner. Personalized federated learning extends this paradigm to overcome heterogeneity acros…
View article: Unified Optimal Transport Framework for Universal Domain Adaptation
Unified Optimal Transport Framework for Universal Domain Adaptation Open
Universal Domain Adaptation (UniDA) aims to transfer knowledge from a source domain to a target domain without any constraints on label sets. Since both domains may hold private classes, identifying target common samples for domain alignme…
View article: Mutual Adaptive Reasoning for Monocular 3D Multi-Person Pose Estimation
Mutual Adaptive Reasoning for Monocular 3D Multi-Person Pose Estimation Open
Inter-person occlusion and depth ambiguity make estimating the 3D poses of monocular multiple persons as camera-centric coordinates a challenging problem. Typical top-down frameworks suffer from high computational redundancy with an additi…
View article: Mn2+-activated dual-wavelength emitting materials toward wearable optical fibre temperature sensor
Mn2+-activated dual-wavelength emitting materials toward wearable optical fibre temperature sensor Open
Photothermal sensing is crucial for the creation of smart wearable devices. However, the discovery of luminescent materials with suitable dual-wavelength emissions is a great challenge for the construction of stable wearable optical fibre …
View article: Learning from Crowds with Sparse and Imbalanced Annotations
Learning from Crowds with Sparse and Imbalanced Annotations Open
Traditional supervised learning requires ground truth labels for the training data, whose collection can be difficult in many cases. Recently, crowdsourcing has established itself as an efficient labeling solution through resorting to non-…
View article: CSD 1791391: Experimental Crystal Structure Determination
CSD 1791391: Experimental Crystal Structure Determination Open
An entry from the Inorganic Crystal Structure Database, the world’s repository for inorganic crystal structures. The entry contains experimental data from a crystal diffraction study. The deposited dataset for this entry is freely availabl…
View article: Aptamer-Conjugated Gold Nanoparticles Targeting Epidermal Growth Factor Receptor Variant III for the Treatment of Glioblastoma
Aptamer-Conjugated Gold Nanoparticles Targeting Epidermal Growth Factor Receptor Variant III for the Treatment of Glioblastoma Open
Li Peng,1,2,* Yanling Liang,1,* Xinxin Zhong,1 Zhiman Liang,1,2 Yinghong Tian,3 Shuji Li,1 Jingxue Liang,4 Ransheng Wang,4 Yuqi Zhong,4 Yusheng Shi,5 Xingmei Zhang1 1Key Laboratory of Mental Health of the Ministry of Education, Guangdong-H…
View article: Long-lived Photon Upconversion Phosphorescence in RbCaF3:Mn2+,Yb3+ and the Dynamic Color Separation Effect
Long-lived Photon Upconversion Phosphorescence in RbCaF3:Mn2+,Yb3+ and the Dynamic Color Separation Effect Open
View article: Evaluation of Seed Set Selection Approaches and Active Learning\n Strategies in Predictive Coding
Evaluation of Seed Set Selection Approaches and Active Learning\n Strategies in Predictive Coding Open
Active learning is a popular methodology in text classification - known in\nthe legal domain as "predictive coding" or "Technology Assisted Review" or\n"TAR" - due to its potential to minimize the required review effort to build\neffective…
View article: Empirical Evaluations of Seed Set Selection Strategies for Predictive Coding
Empirical Evaluations of Seed Set Selection Strategies for Predictive Coding Open
Training documents have a significant impact on the performance of predictive models in the legal domain. Yet, there is limited research that explores the effectiveness of the training document selection strategy - in particular, the strat…