Abstract: A printing system comprises a print fluid deposition assembly, a media transport device, and an air flow control system. The print fluid deposition assembly comprises a carrier plate and a printhead arranged to eject a print fluid through an opening of the carrier plate to a deposition region. The media transport device holds a print medium against the movable support surface by vacuum suction and transports the print medium through the deposition region. The air flow control system comprises an air supply unit comprising air flow guide structure extending into the opening of the carrier plate between the carrier plate and the printhead to flow air through the opening. The air flow control system controls the air supply unit to selectively flow the air based on a location of a print medium relative to the printhead.

Abstract: A method of operating an inkjet printer identifies a thermal load that will be imposed on the printheads in the inkjet printer during execution of a print job and identifies a remedial action to reduce the thermal load if the thermal load increases the temperature of the printheads above an upper threshold of a predetermined temperature range. The thermal load is identified from the ink image content data for the print job, the print job parameters for the print job, and environmental condition measurements received from sensors in the printer. The remedial actions include manipulation of the print job queue, alteration of print job parameters, and active cooling to remove heat from the printheads or media within the printer.

Abstract: A semiconductor structure includes a first shallow trench isolation (STI) structure within a semiconductor substrate. The first STI structure includes a buffer structure, an adhesion structure, an electromagnetic reflection structure, and a fill structure. The adhesion structure is between and adhesively bonded to the buffer structure and the electromagnetic reflection structure. The electromagnetic reflection structure is between the adhesion structure and the fill structure to reflect electromagnetic radiation.

Abstract: A three-dimensional (3D) printer is configured to introduce a liquid metal onto a first part and a second part while the first part and the second part are in contact with one another. The liquid metal subsequently solidifies to join the first part and the second part together to produce an assembly.

Abstract: A method of operating a printer identifies image areas and image features in ink image content data that are likely to produce ink blur within an ink image. The ink image content data for these areas and features is modified to attenuate the appearance of ink blur. The image areas identified as being likely to produce ink blur are the trailing edge and the side edges of the ink image content data. The modification of the ink image content data for the features and predetermined areas improves the quality of the printed ink image.

Abstract: A three-dimensional (3D) metal object manufacturing apparatus is equipped with an orifice cleaning system that removes metal drops that have adhered to a plate, an orifice in the plate, and a nozzle ejecting melted metal drops through the orifice during object forming operations. The orifice cleaning system includes an orifice cleaning tool that consists essentially of a soft carbon material, such as graphite. The orifice cleaning tool is configured with a handle that is gripped by an articulated arm to move the orifice cleaning tool against the plate, the orifice, and a portion of the nozzle at the orifice.

Abstract: This application describes a hardware accelerator, a computer system, and a method for accelerating Graph Neural Network (GNN) node attribute fetching. The hardware accelerator comprises a GNN attribute processor; and a first memory, wherein the GNN attribute processor is configured to: receive a graph node identifier; determine a target memory address within the first memory based on the graph node identifier; determine, based on the received graph node identifier, whether attribute data corresponding to the received graph node identifier is cached in the first memory at the target memory address; and in response to determining that the attribute data is not cached in the first memory: fetch the attribute data from a second memory, and write the fetched attribute data into the first memory at the target memory address.

Abstract: Depositing a seed layer after formation of the MD in order to reduce or prevent epitaxial growth of the seed layer toward the MD. For example, the seed layer may be deposited using CVD and conformal dry etching. In some implementations, the seed layer may be formed of ruthenium (Ru), molybdenum (Mo), or tungsten (W). Accordingly, the seed layer helps reduce or prevent seam formation in the VG, which reduces resistance of the VG by allowing for bottom-up metal growth. Additionally, current leakage from the VG to the MD is reduced or even prevented. As a result, device performance and efficiency are increased and breakdown voltage of the gate structure is also increased. Additionally, because electrical shorts are less likely, yield is increased, which conserves power, raw materials, and processing resources that otherwise would have been consumed during manufacture.

Abstract: A programmable device receives commands from a processor and, based on the commands: identifies a root node in a graph; identifies nodes in the graph that are neighbors of the root node; identifies nodes in the graph that are neighbors of the neighbors; retrieves data associated with the root node; retrieves data associated with at least a subset of the nodes that are neighbors of the root node and that are neighbors of the neighbor nodes; and writes the data that is retrieved into a memory.

Abstract: Systems and methods are provided for automatic speech recognition. In the method, the system obtains a padded sequence by processing a plurality of acoustic signals. The system compresses the padded sequence by reducing the size of the padded sequence to obtain a compressed sequence. The system inputs the compressed sequence into a pre-trained encoder neural network to obtain an encoded sequence and then decompresses the encoded sequence by recovering the encoded sequence to an original sequential ordering. The system inputs the encoded sequence to a decoding module to obtain recognition texts.

Abstract: An amorphous composite solid electrolyte is provided that includes one or more three-dimensional branched macromolecules with a core portion and at least three arm portions connected to the core portion. Each arm portion includes a random copolymer or a block polymer comprising a first monomer and a second monomer with a molar ratio of the first monomer to the second monomer in the range from greater than 0 to less than or equal to 1. An ion conductive electrolytic solution including at least one lithium salt solution in an amount of approximately 1 mol/l to 10 mol/l is entrained within the branched macromolecule, with a weight ratio of the branched macromolecule to the ion conducive electrolytic solution equal to or lower than 1:9, such that the branched macromolecule has a swelling degree of at least 5:1 (liquid:polymer in weight) of the ion conductive electrolytic solution.

Abstract: A flat panel device, such as a flat display, comprises a substrate; a power circuit; a plurality of pixel circuits, each pixel circuit coupled to a pixel-generating device; and a plurality of power lines electrically coupled to the plurality of pixel circuits. The plurality of power lines are also electrically coupled to the power circuit by a plurality of vias traveling through the substrate.

Abstract: A method for processing a plurality of queries is provided according to embodiments of the present disclosure. In this method, based on a plurality of queries and an execution plan for the plurality of quires, a plurality of record identification (ID) numbers can be stored into a pool in a numerical order. Each of the plurality of record ID numbers can identify a data record in a database. Then, the execution plan can be performed to batch a plurality of data records corresponding to the plurality of record ID numbers in the database based on a distribution of the plurality of record ID numbers in the pool.

Abstract: A display comprises a substrate (e.g., glass), a plurality of pixel circuits disposed on a back surface of the substrate, and a plurality of self-emitting devices disposed on a front surface of the substrate. The self-emitting devices are electrically connected to the plurality of pixel circuits by at least one electrically conductive via traveling through the substrate. Each pixel circuit comprises a first and a second transistor and a capacitor. The self-emitting devices may be LEDs or OLEDs for example.

Abstract: A resin for removing phosphorus from water body, and a preparation method therefor and an application thereof. The particle size of the resin is 0.5-0.8 mm; the resin has a porous structure, the specific surface area is 8-25 m2/g, and the pore size distribution is 3-15 nm, the wet apparent density is 0.68-0.74 g/cm3; the wet true density is 1.12-1.18 g/cm3: and the water content of the resin is 43-57% in percentage by weight. The resin is loaded with a functional group having a lanthanum-oxygen bond, so that the resin can selectively adsorb phosphate radicals in the water body. The resin can selectively remove phosphorus in the water body by using a mode of loading lanthanum on weak acid cation resin and utilizing high selectivity of the lanthanum-oxygen bond to phosphate radicals, is easy to resolve and low in synthesis cost, and can be repeatedly used.

Abstract: A multi-sheet cooling buffer suitable for use in a printing device and a method of sheet processing with the cooling buffer are described. The cooling buffer includes an inlet and an outlet and a sheet cooling mechanism. An array of cooling stations, intermediate the inlet and the outlet, are each configured to receive print media sheets independently from the inlet and direct cooled sheets towards the outlet. In the cooling buffer, multiple print media sheets are able to be cooled contemporaneously in respective ones of the cooling stations.

Abstract: A flat panel device, such as a flat display, comprises a substrate; a power circuit; a plurality of pixel circuits, each pixel circuit coupled to a self-emitting device; and a plurality of power lines electrically coupled to the plurality of pixel circuits. The plurality of power lines are also electrically coupled to the power circuit by a plurality of vias traveling through the substrate.

Abstract: A method for fabricating minimal fin length includes the steps of first forming a fin-shaped structure extending along a first direction on a substrate, forming a first single-diffusion break (SDB) trench and a second SDB trench extending along a second direction to divide the fin-shaped structure into a first portion, a second portion, and a third portion, and then performing a fin-cut process to remove the first portion and the third portion.

Abstract: A method for fabricating minimal fin length includes the steps of first forming a fin-shaped structure extending along a first direction on a substrate, forming a first single-diffusion break (SDB) trench and a second SDB trench extending along a second direction to divide the fin-shaped structure into a first portion, a second portion, and a third portion, and then performing a fin-cut process to remove the first portion and the third portion.

Abstract: A method of operating a printer evaluates the effect of printing a print job with two different orientations of the image content data for the print job. The orientation that produces the least increase in the cross-process direction displacement error for the inkjets in the printer is selected for printing the print job. By attenuating the increase in cross-process direction displacement error, the interval between printhead maintenance operations is increased.