Abstract: A semiconductor device structure, along with methods of forming such, are described. The structure includes a fin disposed over a semiconductor substrate, and the fin has a first width. The structure further includes an isolation region disposed around the fin, a gate electrode disposed over the fin and the isolation region, and a fill material disposed in the gate electrode. The fill material is in contact with a top surface of a portion of the semiconductor substrate, the top surface has at least a portion having a substantially flat cross-section, and the portion of the top surface has a second width substantially greater than the first width.

Abstract: The performance of a neural network (NN) and/or deep neural network (DNN) can limited by the number of operations being performed as well as management of data among the various memory components of the NN/DNN. A sparsity-inducing regularization optimization process is performed on a machine learning model to generate a compressed machine learning model. A machine learning model is trained using a first set of training data. A sparsity-inducing regularization optimization process is executed on the machine learning model. Based on the sparsity-inducing regularization optimization process, a compressed machine learning model is received. The compressed machine learning model is executed to generate one or more outputs.

Abstract: A method of removing fluoride ion from waste liquid is provided, which includes providing a calcium source and a plurality of ceramic particles to a waste liquid containing fluoride ion for forming a plurality of calcium fluoride layers wrapping the ceramic particles. The calcium fluoride layers are connected to form a calcium fluoride bulk. The ceramic particles are embedded in the calcium fluoride bulk. The ceramic particles and the calcium fluoride bulk have a weight ratio of 1:4 to 1:20.

Abstract: An electronic shelf label positioning system, an electronic shelf label and a guide rail. The electronic shelf label positioning system includes the electronic shelf label, the guide rail, a PDA and a background server. The electronic shelf label includes a main control SoC, a card reader IC, a screen and a power supply device. The main control SoC is configured to control the screen display and to communicate with an AP. The power supply device is configured to supply power to the electronic shelf label. The guide rail includes a guide rail identification area and a label area. The label area is installed with a plurality of wireless labels each having a unique non-repeated ID number. The guide rail identification area is installed with an identity recognition device, which includes a guide rail ID consisting of the ID numbers of the wireless labels sequentially arranged and summarized.

Abstract: An electronic device and a performance optimization method thereof are provided. The electronic device includes a battery module, a processor, and a controller. The battery module is configured to supply power to the electronic device. The processor has a power limit. The controller is configured to monitor a charging and discharging current of the battery module. In a power connection mode, the controller analyzes a status of the battery module and adjusts the power limit of the processor according to the charging and discharging current.

Abstract: A computer system is provided that includes one or more processors configured to receive user input for inked content to a digital canvas, and process the inked content to determine one or more writing regions. Each writing region includes recognized text and one or more document layout features associated with that writing region. The one or more processors are further configured to tokenize a target writing region of the one or more writing regions into a sequence of tokens, process the sequence of tokens of the target writing region using task extraction subsystem that operates on tokens representing both the recognized text and the one or more document layout features of the target writing region, segment the target writing region into one or more sentence segments, and classify each of the one or more sentence segments as a task sentence or a non-task sentence.

Abstract: Systems and methods for performing spatial line grouping on digital ink stokes. The system includes an electronic processor configured to access a set of hypothetical lines in an electronic document and determine a set of hypothetical line pairings. The electronic processor is also configured to determine, via a gradient boosting tree model, a merge confidence score for each hypothetical line pairing and compare a first merge confidence score with a merge threshold. The first merge confidence score is associated with a first hypothetical line and a first neighboring hypothetical line. The electronic processor is also configured to, in response to the first merge confidence score satisfying the merge threshold, merge the first hypothetical line and the first neighboring hypothetical line to form a first line grouping. The electronic processor is also configured to perform a digital ink stroke analysis on the electronic document based on the first line grouping.

Abstract: Systems and methods for performing spatial line grouping on digital ink stokes. The system includes an electronic processor configured to access a set of hypothetical lines in an electronic document and determine a set of hypothetical line pairings. The electronic processor is also configured to determine, via a gradient boosting tree model, a merge confidence score for each hypothetical line pairing and compare a first merge confidence score with a merge threshold. The first merge confidence score is associated with a first hypothetical line and a first neighboring hypothetical line. The electronic processor is also configured to, in response to the first merge confidence score satisfying the merge threshold, merge the first hypothetical line and the first neighboring hypothetical line to form a first line grouping. The electronic processor is also configured to perform a digital ink stroke analysis on the electronic document based on the first line grouping.

Abstract: A data processing system for performing a semantic analysis of digital ink stroke data implements obtaining the digital ink stroke data representing handwritten text, drawings, or both; analyzing the digital ink stroke data to extract path signature feature information from the digital ink stroke data; analyzing the path signature feature information using a convolutional neural network (CNN) trained to perform a pixel-level sematic analysis of the digital ink stroke data and to output a pixel segmentation map with semantic prediction information for each pixel of digital ink stroke data; analyzing the pixel segmentation map to generate stroke-level semantic information using a pixel-to-stroke conversion model; and processing the digital ink stroke data based on the stroke-level semantic information.

Abstract: A stirrup module comprising an outer stirrup set, an upper-opening stirrup set having an upper opening and provided in an opening of the outer stirrup of the outer stirrup set, and a lower-opening stirrup set having a lower opening and provided in the opening of the outer stirrup of the outer stirrup set. Wherein the upper opening and the opening of the outer stirrup are oriented in the same direction, the lower opening and the upper opening are in opposite directions to each other. Wherein the upper-opening stirrup set has a first bottom of stirrup opposite to the upper opening, and the first bottom of stirrup corresponds to the position of the lower opening; the lower-opening stirrup set has a second bottom rib opposite to the lower opening, and the second bottom rib corresponds to the position of the upper opening.

Abstract: A stirrup module includes first, second, third and third erect bar portions, a bottom rib section. The second erect bar portion is arranged in parallel with the first erect bar portion. The third erect bar portion is arranged in parallel with the first erect bar portion and located between the first and second erect bar portions. The fourth erect bar portion is arranged in parallel with the first erect bar portion and located between the third and the second erect bar portions. A first gap is between the first and third erect bar portions, and a second gap is between the second and fourth erect bar portions. The bottom rib section connects at least portion of the bottom ends of the first, second, third and fourth erect bar portions, and the first and second gaps are each formed with an opening at the end opposite to the bottom rib section.

Abstract: The present invention provides a rebar cage module comprising a first opening rebar net and a second opening rebar net. The first opening rebar net extends from a central direction, and partially forms an elongated groove shape enclosing a first accommodating portion, and has a first opening. The second opening rebar net extends from the central direction, and partially surrounds the central direction to form an elongated groove shape, and has a second opening oriented to the opposite direction with the first opening. Wherein, the first opening rebar net at least partially covers the second opening rebar net, the first opening rebar net is at least partially distributed in the second opening, and the first opening rebar net extends at least partially through the second opening into the second accommodating portion, and making the first accommodating portion and the second accommodating portion at least partially overlap.

Abstract: A yarn made from a mixture of a grain and a first polymer by spinning is provides. The yarn includes a plurality of fibers. Each fiber has a surface layer and a core layer surrounded by the surface layer. The surface layer is made from the grain and the first polymer. The core layer is made from the first polymer. The grain includes a nano-powder mixture and a second polymer. A weight percentage of the nano-powder mixture in the grain is from 60% to 70%. The nano-powder mixture includes silicon dioxide, magnesium oxide and aluminum oxide. A weight ratio of silicon dioxide to magnesium oxide in the nano-powder mixture is from 2:1 to 1:2. A weight ratio of silicon dioxide to aluminum oxide in the nano-powder mixture is from 2:1 to 1:2.

Abstract: Examples pertaining to improvement on user equipment (UE) uplink latency in wireless communications are described. When an apparatus is in a special mode, a processor of the apparatus transmits to a network a request for permission to perform an uplink (UL) transmission for a plurality of times. The processor then receives from the network a grant. In response to receiving the grant, the processor performs the UL transmission to the network. In transmitting the request for the plurality of times, the processor transmits the request for the plurality of times at a frequency higher than a frequency at which the request to perform UL transmissions is transmitted to the network when the apparatus is in a normal operational mode.

Abstract: A communication apparatus includes a radio transceiver, an application processor and a modem processor. The radio transceiver transmits or receives wireless signals to or from a network device in a wireless network to communicate with the network device via a communication link. The application processor is configured to run one or more application programs. The modem processor is coupled to the radio transceiver and the application processor and configured to perform operations including: evaluating a future condition of the communication link according to at least one communication parameter or at least one communication event to obtain an evaluation result; and sending a first indication to the application processor to inform the application processor of the evaluation result. Responsive to reception of the first indication, the application processor is further configured to perform operation including: adapting a codec for multimedia processing of data received from the network device.

Abstract: Methods, systems, and computer program products are provided for signature verification. Signature verification may be provided for target signatures using genuine signatures. A signature verification model pipeline may extract features from a target signature and a genuine signature, encode and submit both to a neural network to generate a similarity score, which may be repeated for each genuine signature. A target signature may be classified as genuine, for example, when one or more similarity scores exceed a genuine threshold. A signature verification model may be updated or calibrated at any time with new genuine signatures. A signature verification model may be implemented with multiple trainable neural networks (e.g., for feature extraction, transformation, encoding, and/or classification).

Abstract: A device detecting system is provided. The device detecting system includes a bar code scanner, a plurality of device accommodating spaces, a screen, and a server. The server obtains bar code information via the bar code scanner and opens one of the device accommodating spaces based on the bar code information to accommodate an electronic device. The server performs a test procedure on the electronic device to generate a test result, and displays the test result and operation information corresponding to the test result on the screen.

Abstract: Methods, systems, and computer program products are provided for signature verification. Signature verification may be provided for target signatures using genuine signatures. A signature verification model pipeline may extract features from a target signature and a genuine signature, encode and submit both to a neural network to generate a similarity score, which may be repeated for each genuine signature. A target signature may be classified as genuine, for example, when one or more similarity scores exceed a genuine threshold. A signature verification model may be updated or calibrated at any time with new genuine signatures. A signature verification model may be implemented with multiple trainable neural networks (e.g., for feature extraction, transformation, encoding, and/or classification).

Abstract: An electronic device is provided. The electronic device includes a knob, a user interface (UI) module, and a human interface device (HID) module. The knob is configured to generate a motion signal. The UI module is configured to generate a setting signal. The HID module includes a first HID definition and a second HID definition stored therein and communicates with the UI module. The HID module interprets the motion signal based on the first HID definition or the second HID definition according to the setting signal to generate input information transferred to the operating system. A control method and a computer program product of the electronic device are also provided.

Abstract: A data processing system for performing a semantic analysis of digital ink stroke data implements obtaining the digital ink stroke data representing handwritten text, drawings, or both; analyzing the digital ink stroke data to extract path signature feature information from the digital ink stroke data; analyzing the path signature feature information using a convolutional neural network (CNN) trained to perform a pixel-level sematic analysis of the digital ink stroke data and to output a pixel segmentation map with semantic prediction information for each pixel of digital ink stroke data; analyzing the pixel segmentation map to generate stroke-level semantic information using a pixel-to-stroke conversion model; and processing the digital ink stroke data based on the stroke-level semantic information.