A5100606554- Navier-Stokes equation solutions
- Advanced Mathematical Physics Problems
- Colorectal Cancer Surgical Treatments
- Colorectal and Anal Carcinomas
- Gas Dynamics and Kinetic Theory
- Cosmology and Gravitation Theories
- Energy Efficient Wireless Sensor Networks
- Complex Systems and Decision Making
- Model Reduction and Neural Networks
- Anorectal Disease Treatments and Outcomes
- Power Systems and Technologies
- Expert finding and Q&A systems
- Model-Driven Software Engineering Techniques
- Product Development and Customization
- Microfluidic and Capillary Electrophoresis Applications
- Simulation Techniques and Applications
- Colorectal Cancer Screening and Detection
- Atomic and Subatomic Physics Research
- Auction Theory and Applications
- Machine Learning and Data Classification
- Manufacturing Process and Optimization
- Advanced Computational Techniques and Applications
- Geometric Analysis and Curvature Flows
- Smart Grid and Power Systems
- Distributed Sensor Networks and Detection Algorithms
Chinese Academy of Sciences
2024-2025
Shanghai University
2024-2025
Shanghai Institute of Optics and Fine Mechanics
2024-2025
University of Chinese Academy of Sciences
2023-2024
Academy of Mathematics and Systems Science
2023-2024
Institute of Mathematics and Informatics
2024
Czech Academy of Sciences, Institute of Mathematics
2024
Nanfang Hospital
2023
Southern Medical University
2023
Fujian Polytechnic of Information Technology
2023
Importance Previous studies have demonstrated the advantages of short-term histopathological outcomes and complications associated with transanal total mesorectal excision (TME) compared laparoscopic TME. However, long-term oncological TME remain ambiguous. This study aims to compare 3-year disease-free survival Objective To evaluate between in patients rectal cancer. Design, Setting, Participants randomized, open-label, noninferiority, phase 3 clinical trial was performed 16 different...
We present a method for active inference with partial observations in stochastic systems through incentive design, also known as the leader-follower game. Consider leader agent who aims to infer follower agent's type given finite set of possible types. Different types followers differ either dynamical model, reward function, or both. assume can partially observe follower's behavior system modeled Markov decision process, which takes an optimal policy maximize total reward. To improve...
Solution-processed top electrodes are crucial for PSC commercialization. This TOC image illustrates solution-processed commercial-scale perovskite photovoltaics.
We prove a continuation criterion for the free boundary problem of three-dimensional incompressible ideal magnetohydrodynamic (MHD) equations in bounded domain, which is analogous to theorem given Beale, Kato, and Majda [Commun. Math. Phys. 94, 61–66 (1984)]. combine energy estimates our previous works [C. Hao T. Luo, Arch. Ration. Mech. Anal. 212(3), 805–847 (2014)] on MHD some Ginsberg [SIAM J. 53, 3366–3384 (2021); arXiv:1811.06154] show that solution can be continued as long curls...
The problem of reward design examines the interaction between a leader and follower, where aims to shape follower's behavior maximize leader's payoff by modifying function. Current approaches rely on an accurate model how follower responds modifications, which can be sensitive modeling inaccuracies. To address this issue sensitivity, we present solution that offers robustness against uncertainties in including 1) breaks ties presence nonunique best responses, 2) inexact knowledge perceives...
Mixture-of-Experts (MoE) has been demonstrated as an efficient method to scale up models. By dynamically and sparsely selecting activated experts, MoE can effectively reduce computational costs. Despite the success, we observe that many tokens in models have uncertain routing results. These nearly equal scores for choosing each expert, demonstrate this uncertainty lead incorrect selections. Inspired by Global Workspace Theory (GWT), propose a new fine-tuning method, GW-MoE, address issue....
This paper focuses on the global well-posedness of Oberbeck–Boussinesq approximation for unsteady motion a drop in another bounded fluid separated by closed interface with surface tension. We assume that initial state is close to ball BR same volume as drop, and boundary small perturbation BR. To begin, we introduce Hanzawa transformation an added barycenter point obtain linearized fixed domain. From there, establish time-weighted estimates solutions shifted equation using maximal Lp–Lq...
The scaling of large language models (LLMs) has revolutionized their capabilities in various tasks, yet this growth must be matched with efficient computational strategies. Mixture-of-Experts (MoE) architecture stands out for its ability to scale model size without significantly increasing training costs. Despite advantages, current MoE often display parameter inefficiency. For instance, a pre-trained MoE-based LLM 52 billion parameters might perform comparably standard 6.7 parameters. Being...
This paper aims to establish the global well-posedness of free boundary problem for incompressible viscous resistive magnetohydrodynamic (MHD) equations. Under framework Lagrangian coordinates, a unique solution exists in half-space provided that norm initial data critical homogeneous Besov space $\dot{B}_{p, 1}^{-1+N/p}(\mathbb{R}_{+}^N)$ is sufficiently small, where $p \in [N, 2N-1)$. Building upon prior work such as (Danchin and Mucha, J. Funct. Anal. 256 (2009) 881--927) (Ogawa Shimizu,...
This paper aims to establish the global well-posedness of free boundary problem for incompressible viscous resistive magnetohydrodynamic (MHD) equations. Under framework Lagrangian coordinates, a unique solution exists in half-space provided that norm initial data critical homogeneous Besov space B _ p , 1 – + N / ( R ) is sufficiently small, where p∈[ N,2 N–1). Building upon prior work such as (Danchin and Mucha, J. Funct. Anal. 256 (2009) 881–927) (Ogawa Shimizu, Differ. Equations 274...
Molecular dynamics simulations have emerged as a fundamental instrument for studying biomolecules. At the same time, it is desirable to perform of collection particles under various conditions in which molecules can fluctuate. In this paper, we explore and adapt soft prompt-based learning method molecular tasks. Our model remarkably generalize unseen out-of-distribution scenarios with limited training data. While our work focuses on temperature test case, versatility approach allows...
Background: Laparoscopic anterior rectal resection (LAR) is a commonly performed surgery for cancer patients. Pelvic floor peritoneum closure (PC), vital procedure in conventional resection, not routinely LAR. Study Design: A total of 1118 consecutive patients with receiving LAR were included this retrospective study. Patients allocated into the PC group and non-PC group. The occurrence postoperative complications was compared between 2 groups. Influential factors anastomotic leakage (AL)...
In order to promote the optimization and improvement of basic information platform Fujian Electric Power, further enhance construction ability GIS for a single map, solve problem online abnormal editing low voltage distribution networks in Fujian, automation, improve efficiency collaborative command operation, distribution, dispatch. Power conducts in-depth research analysis on problems existing map application, system performance, functional application. It is aimed at improving application...