Abstract: The disclosure relates to a 5G or 6G communication system for supporting a higher data transmission rate than a 4G communication system such as LTE. The disclosure relates to a method and an apparatus for enhancement of self-interference channel estimation performance to secure self-interference removal performance in full-duplex (FD) communication and, more specifically, to a method and an apparatus for scheduling a transmission resource for enhanced self-interference channel estimation in the FD communication.

Abstract: A 3D stacked FET may include a back-side wiring layer including a first back-side power line and a second back-side power line, a first FET on the back-side wiring layer, a second FET over the first FET, a front-side wiring layer over the second FET, a first through-electrode connecting the first FET to the second FET, and a second through-electrode connecting the front-side and back-side power lines. The front-side wiring layer may extend in a first direction and may include a front-side power line connected to the second back-side power line. The first FET and the second FET may share a gate extending in a second direction. Each of the first FET and the second FET may include a source and a drain respectively on both sides of the gate in the first direction, and a channel between the source and the drain and surrounded by the gate.

Abstract: Disclosed are systems, apparatuses, processes, and computer-readable media for wireless communications. For example, an example of a process may include receiving, at a network device (e.g., a user equipment (UE)) from a network entity via a number of sensing streams based on a maximum of Nt?J sensing streams, a waveform including communications resources and sensing resources. The waveform has a rank Nt and J is a number of layers scheduled for multiple-input multiple-output (MIMO) communications that is less than Nt. The process may further include processing at least the communications resources of the waveform.

Abstract: A photoresist-removing composition includes a polar organic solvent, an alkyl ammonium hydroxide, an aliphatic amine not including a hydroxy group, and a monovalent alcohol. To manufacture a semiconductor device, a photoresist pattern may be formed on a substrate, and the photoresist-removing composition may then be applied to the photoresist pattern. To manufacture a semiconductor package, a photoresist pattern including a plurality of via holes may be formed on a substrate. A plurality of conductive posts including a metal may be formed inside the plurality of via holes, and the photoresist pattern may be removed by applying a photoresist-removing composition of the inventive concept to the photoresist pattern. A semiconductor chip may be adhered to the substrate between the respective conductive posts.

Abstract: Techniques and systems are provided for generating one or more segmentations masks. For instance, a process may include generating a delta image based on a difference between a current image and a prior image. The process may further include processing, using a transform operation, the delta image and features representing the prior image to generate a transformed feature representation of the prior image. The process may include combining the transformed feature representation of the prior image with features representing the current image to generate a combined feature representation of the current image. The process may further include generating, based on the combined feature representation of the current image, a segmentation mask for the current image.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for regenerative learning to enhance dense predictions. In one example method, an input image is accessed. A dense prediction output is generated based on the input image using a dense prediction machine learning (ML) model, and a regenerated version of the input image is generated. A first loss is generated based on the input image and a corresponding ground truth dense prediction, and a second loss is generated based on the regenerated version of the input image. One or more parameters of the dense prediction ML model are updated based on the first and second losses.

Abstract: Techniques and systems are provided for performing online adaptation of machine learning model(s). For example, a process may include obtaining features extracted from a image by a machine learning model during inference and determining, by the machine learning model based on the features during inference, a plurality of keypoint estimates in the image and/or a bounding region estimate associated with an object in the image. The process may further include generating pseudo-label(s) based on the plurality of keypoint estimates and/or the bounding region estimate. The process may include determining at least one self-supervised loss based on the plurality of keypoint estimates and/or the bounding region estimate. The process may further include adapting, based on the at least one self-supervised loss, parameter(s) of the machine learning model. The process may include generating, using the machine learning model with the adapted parameter(s), a segmentation mask for the image (or another image).

Abstract: Systems and techniques are provided for processing one or more images. For instance, according to some aspects of the disclosure, a method may include obtaining an unlabeled image and generating at least one transformed image based on the unlabeled image. The method may include processing the unlabeled image using a pre-trained semantic segmentation model to generate a first segmentation output. The method may further include processing the at least one transformed image using the pre-trained semantic segmentation model to generate at least a second segmentation output. The method may include fine-tuning, based on the first segmentation output and at least the second segmentation output, one or more parameters of the pre-trained semantic segmentation model.

Abstract: Systems and techniques are provided for processing data (e.g., image data). For instance, according to some aspects of the disclosure, a method may include receiving, at a transformer of a machine learning system, learnable queries, keys, and values obtained from a feature map of a segmentation model of the machine learning system. The method may further include learning, via the transformer, a mapping between an unsupervised output and a supervised output of the segmentation model based on the feature map.

Abstract: A wafer cleaning apparatus includes a base, a roller installation table installed on the base, a wafer support unit disposed at the roller installation table and having a support roller for rotatably supporting an edge of a wafer, a pressing roller installed on the roller installation table and above the wafer support unit, and configured to press opposite surfaces of the wafer, and a driving unit providing a force in a direction, crossing a direction of a central axis of the pressing roller, so that a shape of the pressing roller is deformed. The pressing roller deformed by the driving unit applies a first pressure to a central portion of the wafer and a second pressure, different from the first pressure, to an edge portion of the water.

Abstract: A semiconductor device includes a substrate having a peripheral region and a cell region therein. A first semiconductor active pattern is provided, which protrudes from the substrate in the peripheral region. A second semiconductor active pattern is provided, which protrudes from the substrate in the cell region. A first edge of an upper portion of the first semiconductor active pattern has a rounded shape, and a second edge of an upper portion of the second semiconductor active pattern has a rounded shape. A curvature of the first edge is greater than a curvature of the second edge.

Abstract: Aspects of the present disclosure relate to a novel framework for integrating both semantic and instance contexts for panoptic segmentation. In one example aspect, a method for processing image data includes: processing semantic feature data and instance feature data with a panoptic encoding generator to generate a panoptic encoding; processing the panoptic encoding to generate a panoptic segmentation features; and generating the panoptic segmentation mask based on the panoptic segmentation features.

Abstract: A method for manufacturing a cylindrical battery having multiple tabs is provided in which the cylindrical battery includes an electrode assembly, a cylindrical pouch case, an electrode tab protruding to an upper end of the electrode assembly, and an electrode lead welded and electrically connected to the electrode tab, with the electrode tab being plural. The method includes disposing the plurality of electrode tabs on the electrode assembly; winding the electrode assembly into a cylindrical shape; connecting the electrode lead to the plurality of electrode tabs disposed on the upper end of the electrode assembly; inserting the electrode assembly into the cylindrical pouch case to produce a battery cell; connecting, to the cylindrical pouch case, a gas collecting part for collecting gas generated due to charging and discharging of the electrode; and forming a guide part for preventing deformation of the battery cell in the cylindrical pouch case.

Abstract: Disclosed is a substrate polishing method comprising placing a substrate into a substrate polishing apparatus, rotating each of the substrate and a polishing pad of the substrate polishing apparatus, allowing a bottom surface of the substrate to contact a top surface of the polishing pad, and determining whether the polishing pad would benefit from maintenance. The polishing pad includes a plurality of annular regions that are homocentric with a central point of the top surface of the polishing pad. The step of determining whether the polishing pad would benefit from maintenance includes ascertaining a state of the bottom surface of the substrate, and selecting one of the plurality of annular regions by using information about the state of the bottom surface of the substrate. The one of the plurality of annular regions would benefit from maintenance.

Abstract: A method for manufacturing a cylindrical battery having multiple tabs is provided in which the cylindrical battery includes an electrode assembly, a cylindrical pouch case, an electrode tab protruding to an upper end of the electrode assembly, and an electrode lead welded and electrically connected to the electrode tab, with the electrode tab being plural. The method includes disposing the plurality of electrode tabs on the electrode assembly; winding the electrode assembly into a cylindrical shape; connecting the electrode lead to the plurality of electrode tabs disposed on the upper end of the electrode assembly; inserting the electrode assembly into the cylindrical pouch case to produce a battery cell; connecting, to the cylindrical pouch case, a gas collecting part for collecting gas generated due to charging and discharging of the electrode; and forming a guide part for preventing deformation of the battery cell in the cylindrical pouch case.

Abstract: An organic light emitting display includes a plurality of pixels formed at crossings of data lines and gate line parts and a main power supply line part supplying a cell driving voltage to the pixels. The main power supply line part includes a plurality of main power supply lines disposed along a first direction and main power supply line connection patterns for connecting the adjacent main power supply lines along a second direction substantially perpendicular to the first direction. The main power supply line connection patterns are staggered along the second direction.

Abstract: An organic light-emitting display device is disclosed. The organic light emitting display device includes a display panel including sub-pixels emitting light of at least three colors, and a driver supplying a driving signal to the display panel, wherein each of the sub-pixels emitting at least three colors includes an opening region emitting its own color and a light-emitting participation region additionally emitting the same color as or different color from the its own color.

Abstract: An organic light-emitting display device is disclosed. The organic light emitting display device includes a display panel including sub-pixels emitting light of at least three colors, and a driver supplying a driving signal to the display panel, wherein each of the sub-pixels emitting at least three colors includes an opening region emitting its own color and a light-emitting participation region additionally emitting the same color as or different color from the its own color.

Abstract: A personal media system and method that obtain content playback information of an end device is provided. The personal media system including: an end device reader to, when an end device that collects playback information about content playback is connected, obtain the collected playback information and content information about content stored in the end device; and a mapping unit to map the playback information and the content information, to determine playback information for each content in the end device, and to generate playback information for each content of a user based on the playback information for each content in the end device.

Abstract: An organic light emitting display includes a plurality of pixels formed at crossings of data lines and gate line parts and a main power supply line part supplying a cell driving voltage to the pixels. The main power supply line part includes a plurality of main power supply lines disposed along a first direction and main power supply line connection patterns for connecting the adjacent main power supply lines along a second direction substantially perpendicular to the first direction. The main power supply line connection patterns are staggered along the second direction.