Boe Technology Group Co., Ltd.

Intel Corporation

Beijing Xiaomi Mobile Software Co., Ltd.

Chengdu Boe Optoelectronics Technology Co., Ltd.

Huawei Technologies Co., Limited

Beijing Boe Display Technology Co., Ltd.

Chinese Academy of Sciences

International Business Machines Corporation

Hefei Xinsheng Optoelectronics Technology Co., Ltd.

Other

Microsoft Technology Licensing, LLC

Oracle International Corporation

Tsinghua University

Beijing Boe Optoelectronics Technology Co., Ltd.

Lenovo (beijing) Limited

Tao Wang

Song Zhang

Ningning Li

Tao Sun

Jianguo Li

Yimin Zhang

Wenlong Li

Yangzhou Du

Qiang Li

Wei Wei

Naifu Wu

Kun Wu

Peng Wang

Zhen Wang

Chunyan Xie

Ziran Zhao

Weifeng Zhou

Yurong Chen

David L Graumann

Lakshman Krishnamurthy

Yuan Gao

Yong Xu

Hongbin Liu

Wei Wang

Chuanjun Xia

Liu Liu

Kun Wang

Anbang Yao

Jun Zhang

Yun Qiu

Shangmin Sun

Qi Zhang

Bo Xu

Chao Tian

Peng Cai

Xiang Feng

Lei Wang

Lei Zhang

Zhiqiang Chen

Li Ni Zhang

Yi Zhang

Xin Wang

Yuanjing Li

Yu Wang

Hui Jin

Xi Chen

Jian Li

Hao Wang

Qiang Wang

Zhidong Yuan

Can Yuan

Jifeng Tan

Liang Chen

Yinong Liu

Wei Zhang

Jing Li

Jiantao Liu

Qian Zhang

Jie Zhang

Tao Li

Yaohong Liu

Qiang Zhang

Zhi Zhang

Xiaodong Xie

Jing Min Xu

Peng Li

Cheng Xu

Bin Zhou

Qing Fa Wang

Xiao Lin Sun

Kejun Kang

Jianmin Li

Shiwan Zhao

Wei Hu

Liqiang Chen

Xiaofeng Tong

Yu Hu

Pinfan Wang

Chengyuan Luo

Haifeng Hu

Hua Peng

Jie Liu

Shi Lei Zhang

Chuanxiang Tang

Qiang Yi

Lei Zhao

Dong Hui Liu

Yu Zou

Gang Chen

Yongfeng Xia

Ke Liu

Lin Zhang

Wu Jiang

Chao Xue

Xing Fan

Yansong Li

Sha Liu

Rong Fu

Ya-Ti Peng

Bing Han

Methods, computer program products, and/or systems are provided that perform the following operations: obtaining a target neural network structure and constraints for a target neural network; generating a meta learning network having an associated quantization function based, at least in part, on the target neural network structure; training the meta learning network based, at least in part, on providing a hybrid quantization vector as input to the meta learning network and providing a training dataset to the target neural network; obtaining a plurality of hybrid quantization vectors; determining a new hybrid quantization vector from the plurality of hybrid quantization vectors; and retraining the trained meta learning network based, at least in part, on providing the new hybrid quantization vector as input to the trained meta learning network.

Automatic hybrid quantization for deep neural network

The present disclosure relates to a display panel and a terminal device. The display panel includes a substrate; a driving layer arranged on a side of the substrate; a light-emitting device layer arranged on a side of a pixel circuit area away from the substrate; a first blocking dam arranged on a side of an edge area away from the substrate, the first blocking dam being a ring structure surrounding the peripheral circuit area; a plurality of flow guide strips arranged in a range surrounded by the first blocking dam, each of the flow guide strips extending from the edge area to the peripheral circuit area; and an encapsulation layer including a first inorganic layer, an organic layer and a second inorganic layer. The organic layer is defined in the range surrounded by the first blocking dam.

Display panel including flow guide strips and display device

Discloses are a method for detecting a scheduling deadlock in data query. A query plan tree generated by a query statement includes a common temporary table production operator node, a common temporary table consumption operator node, and a hash join operator node. At least one right table chain in the tree is determined. Operator nodes in the right table chain are traversed from bottom to top starting from the leaf node in the right table chain, and the other operator nodes in the tree are traversed, to analyze an execution order relationship between the operator nodes traversed in the tree. When it is found based on the execution order relationship that there are two identical common temporary table consumption operator nodes in the tree that have scheduling dependency, and their upstream operators are the same common temporary table production operator, it is determined that there is a scheduling deadlock.

Method and apparatus for detecting scheduling deadlock in data query, and device

Disclosed is an organic electroluminescent material and device. The organic electroluminescent material is a novel metal complex that comprises a ligand having a fluorene structure and a ligand having a dibenzofuran, dibenzothiophene, or dibenzoselenophene structure, and has a general formula of M(L)(L)(L), where the ligand Lhas a structure represented by Formula IA, the ligand Lhas a structure represented by Formula IB, and the ligand Lis a monoanionic bidentate ligand. The metal complex may be used as the light-emitting material in the organic electroluminescent device. These novel compounds can greatly improve the device efficiency and provide better device performance. Further provided are an electroluminescent device and a compound formulation.

Organic electroluminescent material and device

An integrated railway movement dispatching and train operation control method, comprising: sensing in-transit information of a signal system; constructing an integrated railway movement dispatching and train operation control minimum system; optimizing a time-saving train driving strategy online; optimizing a multi-train driving strategy in a virtual formation mode; and generating an integrated railway movement dispatching and train operation control adjustment solution.

Integrated railway movement dispatching and train operation control method, system, and application

A display substrate is provided. The display substrate includes: a substrate; a light emitting layer arranged on a side of the substrate; an encapsulation layer arranged on a side of the light emitting layer away from the substrate; and at least two dams arranged around a display area in a non-display area. The encapsulation layer includes an organic material layer, and the dams include a first dam arranged around the display area and a second dam arranged around the first dam, where the first dam includes a first sub-dam arranged discontinuously, a second sub-dam arranged continuously, and at least one third sub-dam connecting the first sub-dam and the second sub-dam; the first sub-dam, the second sub-dam, and the third sub-dam are arranged around to form at least one semi-closed region structure; and at least a part of the organic material layer is located in the semi-closed region structure.

Display substrate and display apparatus

The present disclosure discloses an OLED package structure, a display panel and a method for preparing a package structure. The OLED package structure includes a substrate, an OLED disposed on the substrate, and a reinforcement hole disposed on at least one of the substrate and the OLED and extending along a side of the OLED towards the substrate; and an encapsulation layer covering the OLED, and at least a portion of the encapsulation layer extending into the reinforcement hole. Therefore, the reliability of the package structure can be increased, the adhesive strength of the package layer can be increased, peeling or package failure can be prevented, and the service life of the package structure can be improved.

OLED edge reinforcement holes and forming method

The present application provides a touch substrate, a design structure of a touch electrode layer, a display panel, and a display device. The design structure of the touch electrode layer includes connecting portions and conducting wires. The connecting portions and the conducting wires form a plurality of meshes. One connecting portion is connected with four conducting wires. In conducting wires connected with a same connecting portion, at least one side edge of an end portion of at least one conducting wire connected with the connecting portion is provided with a depression, and a bottom edge of the depression joins an edge of the connecting portion. The touch electrode layer is prepared according to the design structure. The display panel includes the touch electrode layer. The display device includes the display panel.

Touch substrate, design structure of touch electrode layer, display panel, and display device

Provided are an array substrate and a manufacturing method therefor, a display panel, and a display device. The array substrate includes: a plurality of electrodes arranged in an array. The electrode includes a plurality of strip electrodes that are connected to each other and spaced. The plurality of strip electrodes at least have two thicknesses. A distance between two adjacent strip electrodes is less than a preset value. During manufacturing, a mask used for forming each layer of sub-electrodes includes a plurality of strip-shaped shading regions that are connected to each other and spaced in a first direction. Strip-shaped shading regions in a mask used for sub-electrodes formed subsequently are integrally shifted by a first distance in the first direction relative to strip-shaped shading regions in a mask used for sub-electrodes formed last time.

Array substrate and manufacturing method therefor, display panel, and display device

A support structure for a flexible display module is provided. The support structure for the flexible display module includes a plurality of brackets and a connecting assembly. The plurality of brackets are sequentially arranged in a first direction, and extend in a second direction, the second direction intersects the first direction. A connecting assembly is connected to two adjacent brackets, so that the two adjacent brackets are capable of rotating with each other in a set plane. The set plane is perpendicular to the second direction. The connecting assembly is configured to limit an included angle formed between the two adjacent brackets to be within a range of a first angle and a second angle, inclusive.

Support structure for flexible display module, and flexible display device

A training method for a speech translation model, a speech translation method, an apparatus, and a device are provided. The training method includes: after entering a model training phase, controlling the speech translation model to execute a speech translation training task; controlling the speech translation model to simultaneously execute an auxiliary training task of the speech translation training task; and adjusting network parameters of the speech translation model according to the speech translation training task and the auxiliary training task, so as to obtain an updated speech translation model after training.

Training method and apparatus for speech translation model, speech translation method and apparatus, and device

Disclosed are a display substrate and a display apparatus. The display substrate includes: a base substrate, including a display region and a non-display region; an encapsulation dam, disposed on the base substrate and located in the non-display region, and annularly surrounding the display region; and an overflow detection structure, disposed on the base substrate and located in the non-display region, and annularly surrounding the display region, where the overflow detection structure is located between a region where the encapsulation dam is located and the display region. The overflow detection structure includes: at least one convex part; and a reflection part on a side, facing away from the base substrate, of the convex part and at least partially covering the at least one convex part.

A sample mounting device for direct tensile test of rock mass is provided, which includes a first positioning stage, a second positioning stage, a first support piece, a second support piece, and a screw drive mechanism. The second positioning stage is arranged above the first positioning stage and capable of moving up and down, and cushion blocks for gluing with the rock mass are coaxially and detachably arranged at opposite ends of the first positioning stage and the second positioning stage, and a rock mass mounting area is formed between the two cushion blocks. A screw of the screw drive mechanism includes a first thread segment and a second thread segment with opposite thread directions. The first support piece is connected with the first thread segment through a nut seat, and the second support piece is connected with the second thread segment through a nut seat.

Sample mounting device for a direct tensile test of rock mass

The disclosure relates to a display panel, a method for fabricating the same, and a display device. The display panel includes: a base substrate including a first base substrate in a blind hole area and a second base substrate, the thickness of the first base substrate is smaller than the thickness of the second base substrate; an anode layer, a light-emitting layer, a cathode layer and an encapsulation layer arranged successively on the second base substrate, and the cathode layer and the encapsulation layer arranged successively on the first base substrate; wherein at least one installation hole is arranged in a display area of the display panel, the at least one installation hole is within the blind hole area and runs through the first base substrate and the respective layers on the first base substrate.

Display panel, method for fabricating the same, and display device

A metal complex with fluorine substitution is disclosed, which employs a series of new ligands containing fluorine-substituted structure can be used as a luminescent material in an emissive layer of an electroluminescent device. By using the metal complex can provide much longer device lifetime, better thermal stability, no blue-shifted illumination, and higher luminous efficiency. An electroluminescent device and compound formulation are also disclosed.

Metal complex with fluorine substitution

Growing community of inventors

Beijing, China

Tao Wang