Abstract: The present invention discloses a concrete test device and method based on a load-corrosion coupling action, relating to the technical field of corrosion experimental devices for concrete sewage pipelines. The device simulates a real environment of a sewage pipeline based on the load-corrosion coupling action, applies mechanical load and chemical corrosion effects to a concrete specimen simultaneously, and accelerates a corrosion effect on the concrete specimen by exposing the concrete specimen. According to the device, researches on mechanical properties of a concrete material under the load-corrosion coupling action are realized, and an assessment sample basis is provided for researches on aging characteristics of the concrete sewage pipelines. The present invention further discloses a concrete test method based on the aforementioned test device.

Abstract: A method for attacking a neural network that includes receiving an input data that includes an image and ground truth label, adding a pre-determined amount of noise to the image, denoising the noisy image utilizing a diffusion model that includes a deep equilibrium root solver, determining a first gradient of the denoised image with respect to the input data including at least the image, wherein the first gradient is associated with the diffusion model, utilizing the denoised image at downstream model, outputting a predicated label associated with the denoised image, determining a loss utilizing with the predicted label and the ground truth label, determining a second gradient associated with the downstream model utilizing at least the loss, and outputting an aggregate gradient that represents an error of the neural network output utilizing the predicted label, wherein the aggregate gradient is calculated utilizing the first gradient and the second gradient.

Abstract: A system for faster object attribute and/or intent classification may include an machine-learned (ML) architecture that processes temporal sensor data (e.g., multiple instances of sensor data received at different times) and includes a cache in an intermediate layer of the ML architecture. The ML architecture may be capable of classifying an object's intent to enter a roadway, idling near a roadway, or active crossing of a roadway. The ML architecture may additionally or alternatively classify indicator states, such as indications to turn, stop, or the like. Other attributes and/or intentions are discussed herein.

Abstract: A system and method for enhancing a user uplink for satellite communications, including: maintaining a serving reception of the user uplink at a serving satellite; selecting a diversity satellite moving into a competitive position as compared to an elevation angle of the serving satellite; forming a diversity reception of the user uplink at the diversity satellite; and diversity combining the serving reception and the diversity reception to enhance the user uplink for a period longer than a handoff period, wherein the diversity combining is performed while a signal strength difference between the serving reception and the diversity reception is less than or equal to a predefined limit.

Abstract: A camera device comprises an image sensor, a first moving module and two groups of second moving modules. The first moving module is connected to the image sensor. The first moving module is activated to move the image sensor linearly along a first axis. The second moving module is connected to the image sensor. The second moving module is activated to move the image sensor linearly along a second axis, which intersects with the first axis. The two groups of second moving modules are activated simultaneously to rotate the image sensor around a third axis, which is orthogonal to the first and second axes, respectively. The image sensor compensates for the image by linear movement in the direction of the first axis and the second axis and can also compensate for the image by rotation in the direction of the third axis to achieve three-axis image stabilization.

Abstract: A wireless device transmits a multiple consecutive slots transmission (MCSt) via consecutive slots, wherein a quantity of the consecutive slots is based on a channel occupancy time (COT) duration of a COT.

Abstract: Devices, methods and systems for parameterized spatial reuse (PSR). A parameterized spatial reuse reception (PSRR) Physical Layer protocol data unit (PPDU) which includes a trigger frame is received from an access point (AP). A PSR opportunity is identified, responsive the PSRR PPDU, and a PSR-based backoff procedure is started based on the PSR opportunity. A parameterized spatial reuse transmission (PSRT) PPDU is transmitted. Devices, methods and systems for enhanced spatial reuse (ESR). An enhanced spatial reuse reception (ESRR) physical protocol data unit PPDU is received, the ESRR PPDU including a trigger frame that indicates that at least one subchannel may be punctured, and a duration of an ESR operation. An enhanced spatial reuse transmission (ESRT) PPDU is transmitted on at least one punctured subchannel, on an ESR opportunity based on the ESRR PPDU. A response to the ESRT PPDU is received on the at least one punctured subchannel.

Abstract: A computer includes a processor and a memory, the memory storing instructions executable by the processor to receive radar data from an ultrawideband radar in a passenger compartment of a vehicle, determine a starting-time for a time-window based on a second derivative of the radar data, identify a motion inside the passenger compartment based on the radar data received during the time-window, and actuate a component of the vehicle based on the identification of the motion.

Abstract: A method for training a shape optimization neural network to produce an optimized point cloud defining desired shapes of materials with given properties is provided. The method comprises collecting a subject point cloud including points identified by their initial coordinates and material properties and jointly training a first neural network to iteratively modify a shape boundary by changing coordinates of a set of points in the subject point cloud to maximize an objective function and a second neural network to solve for physical fields by satisfying partial differential equations imposed by physics of the different materials of the subject point cloud having a shape produced by the changed coordinates output by the first neural network. The method also comprises outputting optimized coordinates of the set of points in the subject point cloud, produced by the trained first neural network.

Abstract: A method includes forming bottom conductive lines extending along a first direction; forming a first memory cell over one of the bottom conductive lines and a second memory cell over another one of the bottom conductive lines; forming a third memory cell over the one of the bottom conductive lines and a fourth memory cell over the another one of the bottom conductive lines, wherein the third memory cell and the fourth memory cell are at a higher level than the first memory cell and the second memory cell; and forming top conductive lines above the bottom conductive lines, wherein one of the top conductive line vertically overlaps the first and third memory cells, and another one of the top conductive line vertically overlaps the second and fourth memory cells.

Abstract: A patterning process is performed on a semiconductor wafer coated with a bottom layer, a middle layer and a photoresist layer having a starting thickness. The patterning process includes: performing an exposure step including exposing the semiconductor wafer using a mask that includes a feature which produces an intermediate light exposure in a target area followed by processing that creates openings in the photoresist layer in accordance with the mask and thins the photoresist in the target area due to the intermediate light exposure in the target area leaving thinned photoresist in the target area; performing middle layer etching to form openings in the middle layer aligned with the openings in the photoresist layer, wherein the middle layer etching does not remove the middle layer in the target area due to protection provided by the thinned photoresist; and performing trim etching to trim the middle layer in the target area.

Abstract: A dividing method of a wafer includes cutting a front surface of the wafer to a predetermined depth with a cutting blade to form bottomed cut-grooves, applying a water-soluble liquid resin to the front surface of the wafer and drying the water-soluble liquid resin to fill the cut-grooves with the resulting dried water-soluble resin, grinding the wafer at the back surface thereof with grinding stones, so that the cut-grooves are allowed to appear in the back surface of the wafer and the wafer is divided into individual chips, and ejecting rinse water against the wafer, which has been divided into the chips, to remove the water-soluble resin.

Abstract: A network device may originate a route, and may designate the route as a first colored route having a first color. The network device may advertise the first colored route to a first intermediate network device to cause the first intermediate network device to propagate the first colored route to an ingress network device over a first colored border gateway protocol session. The network device may designate the route as a second colored route having a second color, and may advertise the second colored route to a second intermediate network device to cause the second intermediate network device to propagate the second colored route to the ingress network device over a second colored border gateway protocol session.

Abstract: Embodiments of the present application provide a node protection method, device, electronic equipment and medium, which are applied to a first node, comprising: obtaining a first service packet, of which a first SRH extension header contains a first SID list and a preset flag bit, the first SID list containing a primary SID of a primary tail node on a primary path for forwarding the first service packet and a backup SID of a backup tail node, the preset flag bit being set to a first value indicating that the first SRH extension header carries the backup SID; forwarding the first service packet. After a primary tail node is unreachable and before routing converges again, a user connected downstream of the CE device will not experience traffic interruption, which ensures tail node reliability and increases network reliability.

Abstract: A method for performing template-based intra mode derivation (TIMD) with enhanced candidate intra-prediction modes is provided. A video coder determines a set of most probable modes (MPMs) for intra-coding the current block based on intra-prediction modes used to code neighboring blocks of the current block. The video coder determines a template of the current block among already-reconstructed pixels neighboring the current block. The video coder determines a set of candidate intra-prediction modes to include (i) the set of MPMs and (ii) one or more intra-prediction modes neighboring an MPM in the set of MPMs. The video coder derives an intra-prediction mode from the set of candidate intra-prediction modes based on the determined template and the set of candidate intra-prediction modes. The video coder encodes or decodes the current block by using the derived intra-prediction mode to generate an intra-prediction.

Abstract: A pixel array may include a plurality of pixel regions including a first pixel region and a second pixel region. The pixel array may include a metal grid structure over the plurality of pixel regions. The pixel array may include a light blocking layer. A first portion of the light blocking layer may be over the first pixel region and under the metal grid structure. The first portion may have a first thickness. A second portion of the light blocking layer may be over the second pixel region and under the metal grid structure. The second portion may have a second thickness that is different from the first thickness.

Abstract: The present disclosure relates to a method for preparing caronic anhydride. The method comprises cyclizing caronic acid to obtain crude caronic anhydride; and using isopropyl ether to recrystallize the crude caronic anhydride to obtain caronic anhydride.

Abstract: Provided herein are engineered hMPV F proteins. In some aspects, the engineered F proteins exhibit enhanced conformational stability and/or antigenicity. Methods are also provided for use of the engineered F proteins as diagnostics, in screening platforms, and/or in vaccine compositions.

Abstract: The present application provides BCMA targeting chimeric antigen receptor (CAR) comprising a BCMA binding region and an intracellular costimulatory domain derived from DAP10. Further provided are engineered immune effector cells (such as NK cells) comprising the chimeric antigen receptors. Pharmaceutical compositions, kits and methods of treating cancer are also provided.

Abstract: A method of forming a photoresist pattern includes forming an upper layer including a floating additive polymer over a photoresist layer formed on a substrate. The photoresist layer is selectively exposed to actinic radiation. The photoresist layer is developed to form a pattern in the photoresist layer, and the upper layer is removed. The floating additive polymer is a siloxane polymer.