Abstract: Disclosed in an embodiment of the present invention is an electroplating device, comprising an electroplating tank, a clamp, a positioning cylinder and an anode, wherein the positioning cylinder is located in the electroplating tank; the positioning cylinder is open at one end; the anode is located inside the positioning cylinder, and the positioning cylinder comes in contact with the anode in a sealing manner; and in the entire surface region of the anode, only a first surface comes in contact with an electroplating solution, and the first surface is parallel and opposite to a substrate, with the center of the first surface being aligned with the center of the substrate, and the size of the first surface being similar to that of an effective electroplating region of the substrate.

Abstract: The present disclosure provides a display substrate, a method for manufacturing the same, and a display device, which belongs to the technical field of displays. The display substrate includes: a first signal line, located on a base substrate and extending in a first direction; a second signal line, located on the base substrate and extending in a second direction, where the second signal line is arranged in a different layer from the first signal line, the second direction intersects with the first direction, the first signal line includes a first portion, and a first orthographic projection of the first portion on the base substrate is located within a orthographic projection of the second signal line on the base substrate; the display substrate further includes: an isolation pattern, located on a side of the first signal line away from the base substrate.

Abstract: This application discloses a communication method, apparatus, and system, a storage medium, and a program product. The method includes: A first distributed node sends communication mechanism indication information to at least one second distributed node adjacent to the first distributed node, where the communication mechanism indication information is obtained based on a robustness requirement of a distributed learning system; the first distributed node receives at least one piece of first data from the at least one second distributed node, where the at least one piece of first data is transmitted by the at least one second distributed node based on the communication mechanism indication information; and the first distributed node performs inference based on second data of the first distributed node and the at least one piece of first data by using a distributed learning model.

Abstract: A method of correcting perspective distortion of a selfie image captured with a short camera-to-face distance by processing the selfie image and generating an undistorted selfie image appearing to be taken with a longer camera-to-face distance. A pre-trained 3D face GAN processes the selfie image, inverts the 3D face GAN to obtain improved face latent code and camera parameters, fine tunes a 3D face GAN generator, and manipulates camera parameters to render a photorealistic face selfie image. The processed selfie image has less distortion in the forehead, nose, cheek bones, jaw line, chin, lips, eyes, eyebrows, ears, hair, and neck of the face.

Abstract: Disclosed are a winding structure for an electric motor and an electric motor. The electric motor comprises an iron core (1) and a multi-phase winding structure (2), wherein the iron core (1) is provided with a plurality of grooves (11) arranged in the circumferential direction, and one side of the iron core (1) is an inserting side, and the other side is a welding side; the multi-phase winding structure (2) is arranged in the plurality of grooves (11) of the iron core (1), each phase of the winding structure (2) comprises N parallel winding sub-windings formed by a plurality of hairpin conductors of different shapes, and N is a positive even number; and the conductors, forming each phase of the winding structure (2), of the same sub-winding are located on a non-adjacent conductor layer corresponding to the grooves (11).

Abstract: A method for fabricating a film bulk acoustic resonator (FBAR) filter device is provided. The method includes: forming a first electrode of each one of a first resonator and a second resonator on a first surface of a piezoelectric layer, forming a first passivation layer of each one of the first resonator and the second resonator on a corresponding one of the first electrodes, forming a second electrode of each one of the first resonator and the second resonator on a second surface of the piezoelectric layer, conducting a radio frequency (RF) performance test on the FBAR filter device, adjusting a thickness of the second electrode of the first resonator based on a result of the RF performance test, and forming a second passivation layer of each one of the first resonator and the second resonator on a corresponding one of the second electrodes.

Abstract: The invention provides an apparatus for cleaning semiconductor wafers. The apparatus includes at least one first tank, containing cleaning chemical, configured to implement batch cleaning process; one or more second tanks, containing cleaning liquid, configured to implement batch cleaning process; one or more single wafer cleaning modules, configured to implement single wafer cleaning and drying processes; a plurality of robots, configured to transfer one or more wafers to the at least one first tank and the one or more second tanks successively, and then to the one or more single wafer cleaning modules; and a controller, configured to control the plurality of robots to transfer the one or more wafers to the at least one first tank and the one or more second tanks successively, and then to the one or more single wafer cleaning modules.

Abstract: The present disclosure relates to neural network construction methods. One example method includes generating parameters of a target neural network based on a parameter generation network, where input of the parameter generation network includes information about a relative number of a neuron in the target neural network, the relative number of the neuron represents a relative location of the neuron at a first neural network layer, and the first neural network layer is a layer at which the neuron in the target neural network is located. The target neural network is constructed based on the parameters of the target neural network, where the parameters of the target neural network are generated by inputting the information about the relative number of the neuron into the target neural network.

Abstract: Provided is a liquid crystal handwriting board. The liquid crystal handwriting board includes a first substrate, a second substrate, a third substrate, a first liquid crystal layer, and a second liquid crystal layer; wherein the first substrate includes: a first base, and a first pixel electrode and a transparent electrode, wherein the first pixel electrode is insulated from the transparent electrode, and the first pixel electrode is closer to the second substrate than the transparent electrode is; wherein the first substrate, the first liquid crystal layer and the second substrate are configured to form a display portion of the liquid crystal handwriting board; and the first substrate, the second liquid crystal layer and the third substrate are configured to form a writing portion of the liquid crystal handwriting board.

Abstract: The present disclosure provides a display panel and a manufacturing method thereof. The display panel includes a special-shaped display region. The display panel further includes a base substrate and a plurality of pixels on a side of the base substrate, and the plurality of pixels includes an edge pixel with an orthographic projection on the base substrate overlapped with that of an edge of the special-shaped display region on the base substrate. The edge pixel is provided with a light-shielding pattern that divides the edge pixel into a light-shielding region and a light-transmitting region. The edge pixel has a predetermined gray scale and a brightness ratio, the brightness ratio is a ratio of a brightness of the light-transmitting region to a brightness of the edge pixel when an entire region thereof is light-transmitting, and the predetermined gray scale has an exponential function relationship with the brightness ratio.

Abstract: The invention relates to prostate specific membrane antigen humanized antibodies (anti-PSMA) and anti-PSMA antibody drug conjugates. The invention also relates to methods and compositions for using anti-PSMA antibody drug conjugates in inhibiting, preventing or treating PSMA related diseases or cancers.

Abstract: This application relates to a discontinuous reception method. In the first time unit, for example, the first symbol, after a hybrid automatic repeat request HARQ feedback occasion of a sidelink HARQ process, a first terminal may start a drx-HARQ-RTT-TimerSL and start a drx-RetransmissionTimerSL when the drx-HARQ-RTT-TimerSL expires. During running of the drx-RetransmissionTimerSL, the first terminal monitors a physical downlink control channel. In this way, sidelink retransmission efficiency can be improved, an increase of a sidelink retransmission delay can be avoided, and sidelink-based applications, such as unmanned driving, automatic driving, assisted driving, intelligent driving, networked driving, intelligent networked driving, and car sharing in the field of artificial intelligence can be better supported.

Abstract: An example system includes: a production system skill library, with a plurality of skill blocks describing and encapsulating the realization part of the skills involved in the production process; a unified execution engine with a plurality of skill function blocks describing and encapsulating the interface part of the skills involved in the production process; to receive a production procedure programmed by a user based on the skill function blocks, and successively start each skill function block in the production procedure to call at least one corresponding skill block; and device agents for controlling devices in the production system. Each device agent is used to provide a unified interface to control the corresponding device to perform operations according to the operation instructions from the unified execution engine or the skill block.

Abstract: This application provides a machine learning model training method and apparatus. One example method includes: A distributed node sends a first parameter set to a central node. The first parameter set is used to determine a target training method for a machine learning model of the distributed node, and the target training method includes a first training method or a second training method. The central node receives the first parameter set from the distributed node, and determines the target training method for the distributed node based on the first parameter set. The central node sends a first message to the distributed node. The first message includes information about the target training method. The distributed node receives the first message from the central node.

Abstract: The embodiments of the disclosure provides a package structure and a manufacturing method thereof, the package structure includes: a die, including a substrate and a device layer, wherein the substrate has a front side and a back side opposite to each other in a first direction perpendicular to a main surface of the die, and the device layer is at the front side of the substrate; a thermal interface material layer, on the back side of the substrate; and a thermal dissipation component, attached to the die through the thermal interface material layer, the die further includes at least one of a heat dissipation filling structure and a substrate recess, the heat dissipation filling structure is embedded in the substrate and has an exposed part exposed at a sidewall of the substrate; the substrate recess is recessed from the back side of the substrate towards the device layer.

Abstract: This application provides a communication method and a communications device. The method includes: A first device receives a first message sent by a second device. The first message is used to indicate a bearer configuration of a bearer of the first device and an initial status of the bearer configuration, and the bearer configuration includes a first bearer configuration and/or a second bearer configuration. The first device determines the bearer configuration and the initial status of the bearer configuration based on the first message. According to the technical solutions in embodiments of this application, a bearer configuration of a terminal device and an initial status of the bearer configuration of the terminal can be indicated. Because both the bearer configuration of a bearer of the terminal and the initial status of the bearer configuration of the terminal are indicated, communication efficiency is improved.

Abstract: An optical lens assembly with a first optical axis and a second optical axis parallel to the first optical axis, includes a first element group located on an object side of the optical lens assembly; a second element group located on an image side of the optical lens assembly; and a light guide element disposed at the position where the first and second optical axes pass. The first and second element groups being located on the same side of the light guide element. The optical lens assembly has a long focal length, a compact size and a relatively higher illumination when satisfying a specific condition.

Abstract: A storage circuit, a chip, a data processing method, and an electronic device are disclosed. The storage circuit includes: an input control circuit and a memory. The input control circuit is configured to: receive n input data and an input control signal; perform first data processing on the n input data based on the input control signal to obtain n intermediate data corresponding to the n input data one by one; and write the n intermediate data and a sign signal corresponding to the n input data into the memory; the memory is configured to store the n intermediate data and the sign signal; different values of the sign signal respectively represent different processing processes of the first data processing, and n is a positive integer.

Abstract: Embodiments of the present disclosure relate to a coprocessor, a host processor, a crash detection method and an electronic device. In this method, the coprocessor sends a handshake request to the host processor capable of running an operating system. The host processor performs an operation triggered by the handshake request. The host processor sends a positive response to the handshake request to the coprocessor if the operation is performed successfully; and the host processor sends, to the coprocessor, a negative response to the handshake request or no response if the operation is performed successfully. The coprocessor monitors a response to the handshake request from the host processor and determines a detection result of hardware crash associated with the operating system at least based on a monitoring result of this response.

Abstract: An air purifying device includes a purifying apparatus including a hollow purifying shell having an interior forming a purifying cavity. A filtering component configured to filter air is disposed in the purifying cavity. An air supply apparatus includes an air supply shell. An interior of the air supply shell is hollow to form an air supply cavity. An air supply inlet and an air supply outlet communicating with the air supply cavity are formed in the air supply shell. The air supply apparatus is configured to supply air outwards. The air purifying device includes an air guide apparatus that includes an air guide shell. An interior of the air guide shell is hollow to form an air guide cavity. The air guide apparatus is configured to deliver air into the air supply apparatus. An air flow driving apparatus is disposed in the purifying apparatus and/or the air supply apparatus.