Abstract: A bottom-pinned spin-orbit torque magnetic random access memory (SOT-MRAM) is provided in the present invention, including a substrate, a bottom electrode layer on the substrate, a magnetic tunnel junction (MTJ) on the bottom electrode layer, a spin-orbit torque (SOT) layer on the MTJ, a capping layer on the SOT layer, and an injection layer on the capping layer, wherein the injection layer is divided into individual first part and second part, and the first part and the second part are connected respectively with two ends of the capping layer.

Abstract: A layout pattern of a magnetoresistive random access memory (MRAM) includes a substrate having a first cell region, a second cell region, a third cell region, and a fourth cell region and a diffusion region on the substrate extending through the first cell region, the second cell region, the third cell region, and the fourth cell region. Preferably, the diffusion region includes a H-shape according to a top view.

Abstract: A method for evaluating a risk of a subject getting a specific disease includes steps of: storing a reference database that contains original parameter sets; selecting target alleles from an SNP profile derived from genome sequencing data of a subject; selecting target parameter sets from among the original parameter sets; calculating, for each of the target parameter sets, a race factor based on a global risk allele frequency and a group-specific risk allele frequency included in the target parameter set; calculating a genetic factor based on statistics, global reference allele frequencies, the race factors for the target parameter sets, and numbers of chromosomes in homologous chromosome pairs included in the target parameter sets; calculating a citation factor based on numbers of citation times included in the target parameter sets; and calculating a risk score based on the genetic factor and the citation factor.

Abstract: An optical fiber transmission device includes a substrate, a photonic integrated circuit, and an optical fiber assembly. The photonic integrated circuit is disposed on an area of the substrate. The substrate has a protruding structure at an interface with an edge of the photonic integrated circuit. The optical fiber assembly includes an optical fiber and a ferrule that sleeves the optical fiber. The protruding structure of the substrate is configured to abut against the ferrule to limit the position of the optical fiber assembly in a vertical direction of the substrate, such that the protruding structure is a stopper for the optical fiber assembly in the vertical direction.

Abstract: A static auto-tuning system and method for controlling operation of a motor in a system. A speed reference signal is generated resulting in a speed response of the motor. Closed-loop feedback magnifies the rotating friction effect to an observable level. Inertia and rotating friction coefficient values of the system are estimated based on the speed frequency response and a virtual damping coefficient. A fixed low frequency speed signal may result in a first frequency response function for determining virtual damping, and a variable frequency excitation signal may result in a second frequency response function for determining the inertia and rotating friction characteristics. Closed-loop gains are determined based on these characteristics. The excitation signal may be sampled and a peak value in each interval may be identified and stored to produce an envelope of peak values for determining the gain response. Operation of the motor is controlled using the determined gains.

Abstract: An optical fiber network device includes a fiber and a photonic integrated circuit. Fiber receives a first optical signal and transmits a second optical signal. A first wavelength of first optical signal is different from a second wavelength of second optical signal. Photonic integrated circuit includes a laser chip, a photodetector, a wavelength division multiplexing coupler, a first optical modulation element and a second optical modulation element. Laser chip is disposed on photonic integrated circuit, and is configured to generate first optical signal. Photodetector detects second optical signal. Wavelength division multiplexing coupler is configured to couple first optical signal to fiber, and receives second optical signal. First optical modulation element is coupled to wavelength division multiplexing coupler and laser chip, and is configured to modulate first optical signal.

Abstract: A semiconductor device includes a substrate, a stack of semiconductor nanosheets, a dielectric wall, and a gate structure. The substrate includes a nanosheet mesa, and the stack of semiconductor nanosheets is disposed on the nanosheet mesa. The dielectric wall crosses through the nanosheet mesa and the stack of semiconductor nanosheets. The gate structure wraps the stack of semiconductor nanosheets and crosses over the dielectric wall, wherein a top of the dielectric wall has a recess.

Abstract: A method includes forming a pad layer. The pad layer includes a first portion over a first part of a semiconductor substrate, and a second portion over a second part of the semiconductor substrate. The first portion has a first thickness, and the second portion has a second thickness smaller than the first thickness. The semiconductor substrate is then annealed to form a first oxide layer over the first part of the semiconductor substrate, and a second oxide layer over the second part of the semiconductor substrate. The pad layer, the first oxide layer, and the second oxide layer are removed. A semiconductor layer is epitaxially grown over and contacting the first part and the second part of the semiconductor substrate.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for jointly training an image embedding model and a text embedding model. In one aspect, a method comprises: processing data from a historical query log of a search system to generate a candidate set of training examples, wherein each training example comprises: (i) a search query comprising a sequence of one or more words, (ii) an image, and (iii) selection data characterizing how often users selected the image in response to the image being identified by a search result for the search query; selecting a plurality of training examples from the candidate set of training examples; and using the training data to jointly train the image embedding model and the text embedding model.

Abstract: The present invention provides a transparent conductive substrate, sequentially comprising: a first resist layer, a first transparent conductive layer, a transparent core, a second transparent conductive layer, and a second resist layer; wherein the first resist layer is composed of a UV-light sensitive composition (C1); and the second resist layer is composed of a visible-light sensitive composition (C2). The present invention provides a double-side photolithographic method for manufacturing transparent conductive laminates. The transparent conductive laminates manufactured by the inventive method may be incorporated into touch panels.

Abstract: Aspects of the present disclosure provide a method for training a predictor that predicts performance of a plurality of machine learning (ML) models on platforms. For example, the method can include converting each of the ML models into a plurality of instructions or the instructions and a plurality of intermediate representations (IRs). The method can also include simulating execution of the instructions corresponding to each of the ML models on a platform and generating instruction performance reports. Each of the instruction performance reports can be associated with performance of the instructions corresponding to one of the ML models that are executed on the platform. The method can also include training the predictor with the instructions or the IRs as learning features and the instruction performance reports as learning labels, compiling the predictor into a library file, and storing the library file in a storage device.

Abstract: A semiconductor device includes: first and second input circuits in a central region and correspondingly configured to operate in a first voltage domain; first and second single bit level shifters (SBLSs) correspondingly in first and second regions at first and second sides of the central region relative to a first direction and electrically coupled correspondingly to the first and second input circuits, and correspondingly configured to operate in a second voltage domain; and a control circuit configured to toggle each of the first and second SBLSs between a normal state and a standby state when a control signal is received from the control circuit.

Abstract: An electronic device includes a display surface, a back surface with a first portion and a second portion, and a support assembly. The support assembly includes a first, second, and third boards. The first board includes a first surface, detachably covering the first portion, and a second surface. The second board is bendably connected to the first board and combined with the second portion. The third board includes a pivoted end and a free end. The pivoted end is pivotally connected to the second surface and covers the second board and a portion of the first board. When the first board rotates relative to the first portion, the third board also rotates relative to the second board, the second surface faces the third board, and the second board simultaneously moves along the third board. Accordingly, the display surface is raised up a distance relative to the free end.

Abstract: A chip package and method for fabricating the same are provided that includes hybrid bonds between a substrate and integrated circuit devices. In one example, a chip package includes a plurality of integrated circuit (IC) devices mounted on a substrate. The substrate has a die side and a ball side. The die side of the substrate includes a plurality of exposed metal bond pads. Each IC device has a device body. Functional circuitry is formed in the device body, terminating at a plurality of exposed metal bond pads. The plurality of exposed metal bond pads are hybrid bonded to the plurality of exposed metal bond pads.

Abstract: Driver behavior risk assessment and pedestrian awareness may include an receiving an input stream of images of an environment including one or more objects within the environment, estimating an intention of an ego vehicle based on the input stream of images and a temporal recurrent network (TRN), generating a scene representation based on the input stream of images and a graph neural network (GNN), generating a prediction of a situation based on the scene representation and the intention of the ego vehicle, and generating an influenced or non-influenced action determination based on the prediction of the situation and the scene representation.

Abstract: A method of producing a particulate lithium metal oxide or lithium metal phosphate material comprising the steps of providing one or more metal compounds, adding sufficient water to dissolve the one or more metal compounds to form a metal compound solution, adding a first basic solution, a second basic solution and the metal compound solution at predetermined rates to a reaction vessel containing water to form a reaction mixture, heating the reaction mixture while maintaining a pH of the reaction mixture in a predetermined pH range, adding a lithium compound, adding a fatty acid, filtering a precipitate, washing and preferably drying the precipitate, calcining the dried precipitate in an atmosphere containing oxygen to form a calcined lithium metal oxide or lithium metal phosphate, cooling and sizing the calcined lithium metal oxide or lithium metal phosphate to produce a particulate lithium metal oxide or lithium metal phosphate material having a predetermined average particle size.

Abstract: Driver behavior risk assessment and pedestrian awareness may include an receiving an input stream of images of an environment including one or more objects within the environment, estimating an intention of an ego vehicle based on the input stream of images and a temporal recurrent network (TRN), generating a scene representation based on the input stream of images and a graph neural network (GNN), generating a prediction of a situation based on the scene representation and the intention of the ego vehicle, and generating an influenced or non-influenced action determination based on the prediction of the situation and the scene representation.

Abstract: A method includes: forming first, second, and third NWs; forming form first to fourth transistors in corresponding first to fourth groups of active regions, connecting selected transistors amongst the first and second transistors to form first and second input circuits respectively receiving a first input signal in a first domain and a second input signal in the first domain; connecting selected transistors amongst the first and third transistors and amongst the first and fourth transistors to respectively form a first single bit level shifter (SBLS) and a second SBLS; each SBLS operates in the second domain and receives correspondingly versions of the first and second input signals; and connecting selected transistors amongst the first and third transistors to form a control circuit for toggling the first and second SBLSs between a normal and a standby state, a portion of the control circuit and the first SBLS sharing the second NW.

Abstract: A foldable electronic device is disclosed and includes a first body, a second body, a hinge mechanism, and a cover. The hinge mechanism includes a first rack plate disposed to the first body, a second rack plate disposed to the second body, a first gear shaft meshed with the first rack plate, and a second gear shaft meshed with the second rack plate and the first gear shaft. The cover is movably disposed on the hinge mechanism and covers the hinge mechanism. When the second body rotates relative to the first body, the second rack plate, the second gear shaft, the first gear shaft and the first rack plate rotate in sequence, the first rack plate and the second rack plate abut against the cover, so the cover is away from the hinge mechanism and separated from the first body and the second body.

Abstract: A system and method for performing intersection scenario retrieval that includes receiving a video stream of a surrounding environment of an ego vehicle. The system and method also include analyzing the video stream to trim the video stream into video clips of an intersection scene associated with the travel of the ego vehicle. The system and method additionally include annotating the ego vehicle, dynamic objects, and their motion paths that are included within the intersection scene with action units that describe an intersection scenario. The system and method further include retrieving at least one intersection scenario based on a query of an electronic dataset that stores a combination of action units to operably control a presentation of at least one intersection scenario video clip that includes the at least one intersection scenario.