Abstract: A package comprises a substrate including a first surface, and an upper conductive layer arranged on the first surface, a first light-emitting unit arranged on the upper conductive layer, and comprises a first semiconductor layer, a first substrate, a first light-emitting surface and a first side wall, a second light-emitting unit, which is arranged on the upper conductive layer, and comprises a second light-emitting surface and a second side wall, a light-transmitting layer arranged on the first surface and covers the upper conductive layer, the first light-emitting unit, and the second light-emitting unit, a light-absorbing layer, which is arranged between the substrate and the light-transmitting layer in a continuous configuration of separating the first light-emitting unit and the second light-emitting unit from each other, and a reflective wall arranged on the first side wall, wherein a height of the reflective wall is lower than that of the light-absorbing layer.

Abstract: Embodiments of this application disclose a call method and an apparatus. In the call method, when a user does not actively select an audio device as a voice pickup device and a voice play device, after establishing a call connection to another electronic device, an electronic device selects, from available audio devices, an audio device that meets a user expectation as the voice pickup device and the voice play device. According to technical solutions provided in the embodiments of this application, user experience in a call process can be improved.

Abstract: Disclosed is a method for enhancing memory utilization and throughput of a computing platform in training a deep neural network (DNN). The critical features of the method includes: calculating a memory size for every operation in a computational graph, storing the operations in the computational graph in multiple groups with the operations in each group being executable in parallel and a total memory size less than a memory threshold of a computational device, sequentially selecting a group and updating a prefetched group buffer, and simultaneously executing the group and prefetching data for a group in the prefetched group buffer to the corresponding computational device when the prefetched group buffer is update. Because of group execution and data prefetch, the memory utilization is optimized and the throughput is significantly increased to eliminate issues of out-of-memory and thrashing.

Abstract: A manufacturing method of a semiconductor super-junction device includes the following steps: An n-type substrate is etched in a self-aligning manner using a first insulating layer and a second insulating layer as a mask to form a second groove in the n-type substrate. A gate structure is formed in the second groove.

Abstract: Systems and methods of analyzing on-shelf price tag labels and products at a product storage facility include an image capture device that captures one or more images of one or more product storage structures at a product storage facility. A computing device communicatively coupled to the image capture device analyzes the images of the product storage structures captured by the image capture device and detects individual price tag labels and individual products located on the product storage structure. Based on the detection of the price tag labels and the products, the computing device also defines separate product storage spaces of the product storage structure, determines which price tag labels are allocated to which of the separate product storage spaces, and associates in a database the price tag labels allocated to the product storage spaces with the products stored in those product storage spaces.

Abstract: An embodiment of the disclosure provides an electronic assembly including a stacked structure, a first integrated circuit, a first passive component, and a first electrode. The stacked structure comprises a plurality of insulating layers and a plurality of conductive layers. The first passive component is disposed between the stacked structure and the first integrated circuit. The first electrode is disposed between the stacked structure and the first passive component. The first passive component is electrically connected to the stacked structure through the first electrode.

Abstract: Example methods, apparatuses, and computer-readable media for offline transaction processing are disclosed.

Abstract: An inflatable active packer includes: an inner steel pile; and a sleeve coaxially disposed outside the inner steel pile. The inner diameter of the sleeve is larger than the outer diameter of the inner steel pile. The length of the sleeve is less than or equal to the length of the inner steel pile. The sleeve includes an inner wall, and one end of the inner wall is provided with a guide lug. A strengthened clamping layer is attached to the inner wall of the sleeve. A rubber capsule is fixedly disposed in the middle of the strengthened clamping layer; two ends of the strengthened clamping layer are fixedly connected to the sleeve through a compression member. A mounting hole is disposed in the middle of the strengthened clamping layer. The sleeve includes a through hole coaxially with the mounting hole.

Abstract: A system includes a first device and a second device. The first device is configured to generate a target processing task based on a target object, generate a first identification code corresponding to the target object based on the target processing task, and display the first identification code; the second device is configured to obtain the first identification code through scanning, generate a task execution instruction corresponding to the target processing task based on the first identification code when first verification on the first identification code succeeds, establish a Bluetooth connection channel with the first device, and send the task execution instruction to the first device based on the Bluetooth connection channel; and the first device is further configured to receive the task execution instruction, and process the target processing task based on the task execution instruction when second verification on the task execution instruction succeeds.

Abstract: The present application provides a semiconductor device and the method of making the same. The method includes recessing a fin extending from a substrate, forming a base epitaxial feature on the recessed fin, forming a bar-like epitaxial feature on the base epitaxial feature, and forming a conformal epitaxial feature on the bar-like epitaxial feature. The forming of the bar-like epitaxial feature includes in-situ doping the bar-like epitaxial feature with an n-type dopant at a first doping concentration. The forming of the conformal epitaxial feature includes in-situ doping the conformal epitaxial feature with a second doping concentration greater than the first doping concentration.

Abstract: At least some embodiments of the present disclosure relate to an electronic package structure. The electronic package structure includes an electronic structure, a wiring structure disposed over the electronic structure, a bonding element connecting the wiring structure and the electronic structure, and a reinforcement element attached to the wiring structure. An elevation difference between a highest point and a lowest point of a surface of the wiring structure facing the electronic structure is less than a height of the bonding element.

Abstract: A consumable is provided. The consumable includes (a) at least one sweetener, and (b) a sweetness modifier comprising an extract or fraction from Gnaphalium affine or at least one flavonoid compound. The at least one sweetener is present is a sweetening amount.

Abstract: Disclosed is a modeling simulation system for plasma control, comprising: an interactive modeling environment module and a parameter configuration component, configured to represent and construct a plasma control model and a response model; a simulation engine, configured to invoke a calculation code of a simulation element or perform a simulation calculation by using a solver, and transmit calculated simulation results back to users; a system library, comprising a basic library and a customization library; a device manager, configured to extend a communication interface by using a simulation element to support a communication between models and external systems and guarantee message synchronization.

Abstract: Systems and methods of monitoring inventory of a product storage facility include an image capture device configured to move about the product storage areas of the product storage facility and capture images of the product storage areas from various angles. A computing device coupled to the image capture device obtains the images of the product storage areas captured by the image capture device and processes the obtained images of the product storage areas to detect individual products captured in the obtained images. Based on detection of the individual products captured in the images, the computing device analyzes each of the obtained images to detect one or more adjacent product storage structures (shelves, pallets, etc.) and identifies and selects a single image that fully shows a product storage structure of interest and fully shows each of the products stored on the product storage structure of interest.

Abstract: A method and apparatus for hydrocracking mineralized refuse pyrolysis oil. The method may use the following steps: (a) crushing and pyrolyzing mineralized refuse to obtain arene and alkane precursor biomass oil; (b) hydrogenating the arene and alkane precursor biomass oil obtained in step (a), and separating the obtained hydrocrackate to obtain arene and alkane; and (c) purifying, recovering and optimizing the arene and alkane obtained in step (b), and performing deep processing to produce naphtha, jet fuel, light diesel oil, and heavy diesel oil.

Abstract: An automated driving system includes an object detection system. A neural network image encoder generates image embeddings associated with an image including an object. A neural network text encoder generates concept embeddings associated with each of a plurality of concepts. Each of the plurality of concepts is associated with one of at least two object classes. A confidence score module generates a confidence score for each of the plurality of concepts based on the image embeddings and the concept embeddings associated with the concept. An object class prediction module generates a predicted object class of the object based on an association between a set of concepts of the plurality of concepts having at least two of the highest values of the generated confidence scores and the one of the at least two object classes associated with a majority of the set of concepts.

Abstract: In some embodiments, the present disclosure relates to a device having a semiconductor substrate including a frontside and a backside. On the frontside of the semiconductor substrate are a first source/drain region and a second source/drain region. A gate electrode is arranged on the frontside of the semiconductor substrate and includes a horizontal portion, a first vertical portion, and a second vertical portion. The horizontal portion is arranged over the frontside of the semiconductor substrate and between the first and second source/drain regions. The first vertical portion extends from the frontside towards the backside of the semiconductor substrate and contacts the horizontal portion of the gate electrode structure. The second vertical portion extends from the frontside towards the backside of the semiconductor substrate, contacts the horizontal portion of the gate electrode structure, and is separated from the first vertical portion by a channel region of the substrate.

Abstract: Disclosed is a method for enhancing memory utilization and throughput of a computing platform in training a deep neural network (DNN). The critical features of the method includes: calculating a memory size for every operation in a computational graph, storing the operations in the computational graph in multiple groups with the operations in each group being executable in parallel and a total memory size less than a memory threshold of a computational device, sequentially selecting a group and updating a prefetched group buffer, and simultaneously executing the group and prefetching data for a group in the prefetched group buffer to the corresponding computational device when the prefetched group buffer is update. Because of group execution and data prefetch, the memory utilization is optimized and the throughput is significantly increased to eliminate issues of out-of-memory and thrashing.

Abstract: This application provides novel bispecific antigen-binding antibodies, and antigen-binding fragments thereof, that bind to both HER2 and Trop-2. These bispecific anti-HER2/Trop-2 antibodies comprise a first antigen-binding domain that specifically binds human HER2 or Trop-2, and a second antigen-binding domain that specifically binds human Trop-2 or HER2. The bispecific antibodies have high binding affinity for HER2 and Trop-2, can be internalized by cells expressing HER2 and/or Trop-2. The bispecific antibody-drug conjugates (ADC) are capable of inhibiting the growth of tumors cells expressing HER2 and/or Trop-2 in vitro and in vivo. The bispecific antibodies can be used to diagnose, prognose, and treat HER2- and/or Trop-2-associated human diseases (e.g., cancer, infectious diseases, autoimmune diseases, asthma, transplant rejection, and inflammatory disorders).

Abstract: Composite particles of the present invention include alumina particles and an inorganic coating disposed on a surface of the alumina particles, the alumina particles containing molybdenum (Mo), the inorganic coating including a composite metal oxide.