Yijiang Li

ORCID: 0000-0003-0644-7199
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Minerals Flotation and Separation Techniques
  • Mineral Processing and Grinding
  • Domain Adaptation and Few-Shot Learning
  • Advanced Neural Network Applications
  • Osteoarthritis Treatment and Mechanisms
  • Adversarial Robustness in Machine Learning
  • Coal and Its By-products
  • Diamond and Carbon-based Materials Research
  • Organ Donation and Transplantation
  • Privacy-Preserving Technologies in Data
  • Multimodal Machine Learning Applications
  • Topic Modeling
  • Periodontal Regeneration and Treatments
  • Renal Transplantation Outcomes and Treatments
  • Metallurgical Processes and Thermodynamics
  • Metal Extraction and Bioleaching
  • Hydrocarbon exploration and reservoir analysis
  • Metal and Thin Film Mechanics
  • Nonlinear Optical Materials Studies
  • Advanced Image and Video Retrieval Techniques
  • Coal Properties and Utilization
  • Silk-based biomaterials and applications
  • Proteoglycans and glycosaminoglycans research
  • Coagulation and Flocculation Studies
  • Clay minerals and soil interactions

University of California, San Diego
2025

Xidian University
2023-2025

China University of Mining and Technology
2018-2024

Johns Hopkins University
2023-2024

Soochow University
2022-2024

Wuhu Institute of Technology
2024

Taiyuan University of Science and Technology
2024

Hunan University
2022-2023

Institute of Coal Chemistry
2023

Zhangzhou Municipal Hospital of Fujian Province
2023

In this study, we dive deep into the inconsistency of pseudo targets in semi-supervised object detection (SSOD). Our core observation is that oscillating pseudo-targets undermine training an accurate detector. It injects noise student's training, leading to severe overfitting problems. Therefore, propose a systematic solution, termed Consistent-Teacher, reduce inconsistency. First, adaptive anchor assignment (ASA) substitutes static IoU-based strategy, which enables student network be...

10.1109/cvpr52729.2023.00316 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023-06-01

This article investigates the potential usage of eggshell powder as a bio-modifier for asphalt binder and mixture with consideration workability mechanical properties. The effects varying dosages (3%, 6%, 9%, 12%) on workability, rutting, moisture damage were investigated. Viscosity, multiple stress creep recovery, bitumen bond strength tests performed binders. Subsequently, Superpave gyratory densification indices (Locking point method), wheel tracking, indirect tensile (Modified Lottman)...

10.1080/10298436.2021.1905809 article EN International Journal of Pavement Engineering 2021-04-12

Deep co-training has been introduced to semi-supervised segmentation and achieves impressive results, yet few studies have explored the working mechanism behind it. In this work, we revisit core assumption that supports co-training: multiple compatible conditionally independent views. By theoretically deriving generalization upper bound, prove prediction similarity between two models negatively impacts model's ability. However, most current are tightly coupled together violate assumption....

10.1109/iccv51070.2023.01471 article EN 2021 IEEE/CVF International Conference on Computer Vision (ICCV) 2023-10-01

In mineral processing operations, it is essential to effectively remove the fine particles in slime and tailings slurry with appropriate methods. this study, a unique two-stage flocculation scheme was developed for achieving effective separation of kaolinite from water by polyacrylamide. The turbidity kinetic parameters were used characterize effect dosing ratio between first second stage on behavior settling rate. results showed that maximum rate minimum supernatant achieved when 60% 40%...

10.1080/08827508.2024.2312526 article EN Mineral Processing and Extractive Metallurgy Review 2024-02-04

With a dense two-dimensional hole gas (2DHG) p-type conductive layer near the surface, hydrogen-terminated (C–H) diamond metal-oxide-semiconductor field-effect transistors (MOSFETs) have shown typical normally-on operations and high breakdown voltages ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">V</i> <sub xmlns:xlink="http://www.w3.org/1999/xlink">BR</sub> ). Owing to MOS interface quality, oxidized silicon-terminated (C–Si–O) MOSFETs...

10.1109/led.2025.3528120 article EN IEEE Electron Device Letters 2025-01-01

Different forms of biomaterials, including microspheres, sponges, hydrogels, and nanofibers, have been broadly used in cartilage regeneration; however, effects internal structures the biomaterials on cells chondrogenesis remain largely unexplored. We hypothesized that different sponges hydrogels led to phenotypic disparity may lead disparate chondrogenesis. In current study, chondrocytes chitosan were compared with regard cell distribution, morphology, gene expression, production...

10.1089/ten.tea.2012.0393 article EN Tissue Engineering Part A 2013-04-24

Scaffold incorporated with affinity peptides can efficiently promote cartilage regeneration without exogenous addition of growth factors and cells.

10.1039/c7tb02132a article EN Journal of Materials Chemistry B 2018-01-01

Dataset distillation (DD) allows datasets to be distilled fractions of their original size while preserving the rich distributional information so that models trained on can achieve a comparable accuracy saving significant computational loads. Recent research in this area has been focusing improving datasets. In paper, we aim explore new perspective DD. We study how embed adversarial robustness datasets, these maintain high and meanwhile acquire better robustness. propose method achieves...

10.1609/aaai.v39i9.32978 article EN Proceedings of the AAAI Conference on Artificial Intelligence 2025-04-11

Biodegradable polymers, either as porous scaffolds or microspheres, have been investigated broadly for cartilage tissue engineering. A combination of these two forms materials could potentially maximize their benefits. In this study, poly (l-lactide-co-ε-caprolactone) (PLCL) were integrated with (β-hydroxybutyrate-co-β-hydroxyvalerate) (PHBV) microspheres to enhance the mechanical properties well regulate cell behavior through altering surface topography. PHBV fabricated an emulsion solvent...

10.1088/1748-6041/8/2/025005 article EN Biomedical Materials 2013-02-05

Kidney transplantation is typically the most effective treatment for patients with end-stage renal disease. However, supply of kidneys far short fast-growing demand. paired donation (KPD) programs provide an innovative approach increasing number available kidneys. In a KPD program, willing but incompatible donor-candidate pairs may exchange donor organs to achieve mutual benefit. Recently, research on exchanges initiated by altruistic donors (ADs) has attracted great attention because...

10.1109/tbme.2012.2195663 article EN IEEE Transactions on Biomedical Engineering 2012-04-23
Coming Soon ...