Abstract: In a method of manufacturing a semiconductor device, an underlying structure is formed over a substrate. A film is formed over the underlying structure. Surface topography of the film is measured and the surface topography is stored as topography data. A local etching is performed by using directional etching and scanning the substrate so that an entire surface of the film is subjected to the directional etching. A plasma beam intensity of the directional etching is adjusted according to the topography data.

Abstract: A memory device includes a memory array comprising a plurality of memory cells arranged over a plurality of rows, the rows including a plurality of word lines, respectively, a first group of the memory cells coupled to an even-numbered one of the word lines and a second group of the memory cells coupled to an odd-numbered one of the word lines. The even-numbered word line is disposed in a first one of a plurality of metallization layers formed vertically above a substrate, wherein the even-numbered word line extends along a first lateral direction and includes a first stitch portion extending in a second lateral direction perpendicular to the first lateral direction. The odd-numbered word line is disposed in a second one of the plurality of metallization layers, wherein the odd-numbered word line extends along the first lateral direction and includes a second stitch portion extending in the second lateral direction.

Abstract: Methods and apparatus are disclosed for subnetwork operations. A method comprises, obtaining at an apparatus, a radio network identity which identifies the apparatus as a wireless communication device within a radio network: obtaining a physical layer subnet-work identity, PSI, at least based on the obtained radio network identity: and generating one or more subnetwork signals based on the obtained PSI, wherein the one or more subnetwork signals are used for intra-subnetwork communication within a subnetwork of the radio network.

Abstract: A method may include: sharing, by a client computer program and a server computer program, a set of identification keys, each identification key associated with a key label, and an authentication key; selecting, by the client computer program and the server computer program, one of the key labels; preparing, by the client computer program, quantum systems using a basis, randomly chosen bit values, and intensities; sending, by the client computer program, the quantum systems to the server computer program over a quantum communication channel, wherein the server computer program may be configured to measure the quantum systems using the basis and to announce quantum systems with photon detection; and generating, by the client computer program, a client tag using a shared keyed hash function executed on the authentication key and chosen bit values from the quantum systems with photon detection, and forwarding the client tag to the server computer program.

Abstract: A heat dissipation device and a computer device are disclosed. The heat dissipation device includes a case and a fan module. The case includes a first wall, a second wall and an accommodation cavity. The accommodation cavity is located between the first wall and the second wall in a first direction. The first wall includes a first through hole. The second wall includes a second through hole. The first through hole is connected to the accommodation cavity, the second through hole is connected to the accommodation cavity. The fan module is mounted in the accommodation cavity. The fan module includes a bracket assembly and a plurality of fans. The bracket assembly is detachably connected to the plurality of fans and the case.

Abstract: A user prompt, such as a user prompt received by a client device and sent to an online game system, is provided into a trained large language model. The large language model identifies keywords corresponding to the user prompt. These keywords may be provided to a search engine that identifies corresponding object(s) to place in a virtual experience. The large language model further processes the user prompt to determine spatial placement information for the objects and places the objects accordingly. Subsequently, the system may iteratively receive more prompts and update the virtual experience based on the additional prompts. The placement may be facilitated using macros. The prompts may also affect other attributes of the objects. The knowledge built-into the LLM allows it to suggest which objects are relevant and what quantity and arrangements of these objects is consistent with a scene requested in the user prompt.

Abstract: This application provides an image display method for an electronic device, an encoding method, and a related apparatus. The image display method for an electronic device includes: obtaining a to-be-processed image; obtaining high-luminance enhancement data, where the high-luminance enhancement data includes a high dynamic range layer (hdrLayer); obtaining initial backlight luminance of the electronic device; obtaining target backlight luminance of the electronic device based on the initial backlight luminance; performing luminance adjustment on the to-be-processed image based on the hdrLayer to obtain a target image for the target backlight luminance; and displaying the target image at the target backlight luminance. In this application, a luminance range of a screen can be fully utilized for image display, and optimal end-to-end experience is achieved.

Abstract: A ring buffer storage method includes generating data of a first output according to Q input tokens of a large language model (LLM), and writing the data of the first output into last Q column vectors of an updated first cache tensor buffer matrix. A starting memory address of a first cache tensor buffer is shifted according to the number Q of input tokens of the LLM for updating the first cache tensor buffer. The first cache tensor buffer forms a first cache tensor buffer matrix. The updated first cache tensor buffer forms the updated first cache tensor buffer matrix. The first cache tensor buffer matrix includes a plurality of space segments. Each row of the first cache tensor buffer matrix includes C space segments. C is a cache size. The plurality of space segments have continuous memory addresses.

Abstract: A method of operating a display includes performing a non-synchronized touch scan pattern on a display with a controller coupled to the display. The non-synchronized touch scan pattern schedules touch scans independent of a refresh rate of the display. Upon the controller detecting a first synchronization pulse from a display controller coupled to the controller and the display, a first pulse-checking timer is started. Upon detecting a second synchronization pulse from the display controller and before the first pulse-checking timer expires, a first display refresh rate for the display is obtained from an interval between the first synchronization pulse and the second synchronization pulse. A synchronized touch scan pattern is performed with the controller, and is scheduled to avoid touch scans coinciding with refreshes of the display performed at the first display refresh rate.

Abstract: An anti-twist structure of a voice coil motor includes a base, a lens housing, a first elastic sheet, a second elastic sheet, a magnet, and a yoke member. The lens housing has first margin wall and a second margin wall, and a first protrusion extends from the first margin wall. The height of the second protrusion is lower than the height of the first protrusion. The yoke member has a first wall, a connection wall, a second wall, and a side wall. The first wall is disposed above the first protrusion, and the second wall is disposed above the second protrusion. The lens housing has a deflectable angle relative to a horizontal reference line. When the lens housing deflects to a maximum value of the deflectable angle, the first margin wall abuts against the second wall and/or the first protrusion abuts against the first wall.

Abstract: A manufacturing method of an electronic package includes the following steps. A first interfacial dielectric layer is formed to cover sides of multiple first conductive vias and multiple second conductive vias. Multiple chips are directly bonded to the first and second conductive vias. A base dielectric layer is formed to fill a gap between the adjacent chips. A bridge element is directly bonded to the first conductive vias, such that the bridge element partially overlaps the adjacent chips respectively. A second interfacial dielectric layer and multiple third conductive vias are formed on the first interfacial dielectric layer and the bridge element. A redistribution circuit structure is formed on the second interfacial dielectric layer and the third conductive vias. Multiple conductive bumps are formed on the redistribution circuit structure. An electronic package is also provided.

Abstract: A package structure is provided. The package structure includes a substrate, a first electronic component, a first electrical connector, and a protective layer. The first electronic component is over the substrate. The first electrical connector is between the substrate and the first electronic component. The protective layer encapsulates the first electrical connector. The protective layer has a first curved lateral surface concave toward the first electrical connector and recessed with respect to a lateral surface of the first electronic component.

Abstract: A package assembly and a manufacturing method thereof are provided. The package assembly includes a photonic integrated circuit component, an electric integrated circuit component, a lens and an optical signal port. The photonic integrated circuit component comprises an optical input/output portion configured to transmit and receive optical signal. The electric integrated circuit component is electrically connected to the photonic integrated circuit component. The lens is disposed on a sidewall of the photonic integrated circuit component. The optical signal port is optically coupled to the optical input/output portion.

Abstract: An integrated air valve structure includes a main body and two air valves. The main body is formed with two air passages not communicated with each other, a plurality of through holes disposed corresponding to the two air passages, and two valve mounting seats. One valve mounting seat is disposed corresponding to one of the through holes, the other valve mounting seat is disposed corresponding to two of the through holes belonging to the two air passages. The two air valves are disposed on the two valve mounting seats, each air valves includes an air plug facing at least one of the through holes, a valve body assembled with one of the two valve mounting seats for the air plug to move therein, and a coil disposed on the valve body for generating magnetic force to change a position of the air plug.

Abstract: A power supply system includes a voltage conversion circuit and at least one voltage detector. The voltage conversion circuit converts AC voltage into DC voltage. The voltage detector includes a voltage dividing circuit, a phase voltage detection circuit, and a line voltage detection circuit. The voltage dividing circuit is coupled to the voltage conversion circuit to receive AC voltage. The voltage dividing circuit includes multiple impedance elements and multiple voltage dividing nodes to output multiple divided voltages. The phase voltage detection circuit is coupled to one of the voltage dividing nodes of voltage dividing circuit to generate a phase voltage detection signal based on one of the divided voltages. The line voltage detection circuit is coupled to a part of the voltage dividing nodes of voltage dividing circuit to generate a line voltage detection signal based on a part of the divided voltages.

Abstract: Computer implemented method for converting PDF documents into human readable and machine parsable HTML code. The method includes the use of a machine learning algorithm in order to automatically annotate the HTML code, said algorithm being trained with a set of manually annotated HTML code examples.

Abstract: A Lentinula edodes-derived saltiness enhancing peptide, a method for preparing the same and use thereof are provided herein, belonging to the technical field of active peptides. The Lentinula edodes-derived saltiness enhancing peptide is derived from an enzymolysis extract of a Lentinula edodes fruiting body by flavourzyme and shows a strong ability to enhance saltiness. Compared with the saltiness value of a 4.0 g/L NaCl solution, the Lentinula edodes-derived saltiness enhancing peptide could replace about 50% of NaCl, thus reducing salt without reducing saltiness. Moreover, the Lentinula edodes-derived saltiness enhancing peptide also has a high umami value and can be used in preparation of low-salt foods including low-salt and flavor-enhancing foods.

Abstract: Disclosed in the present invention are a high-strength steel composite galvanized sheet resistant to LME cracking during spot welding, and a preparation method therefor. The high-strength steel composite galvanized sheet comprises a high-strength steel body (1), low-carbon steel composite layers (2) and a galvanized layer (3), wherein two low-carbon steel composite layers (2) are respectively compounded and rolled on two surfaces of the high-strength steel body (1), the galvanized layer (3) is formed on the surface of at least one low-carbon steel composite layer (2), and a high-strength steel composite galvanized sheet is thus formed.

Abstract: The present disclosure relates generally to terpene glycosides, such as certain such compounds extracted from Stevia rebaudiana Bertoni, Rubus suavissimus, or Siraitia grosvenorii. The disclosure also provides for the use of such compounds as food ingredients, flavors, and sweeteners, and related methods. The disclosure also provides ingestible compositions comprising such compounds, as well as processes for extracting such compounds selectively from certain plant sources.

Abstract: An intelligent waste wooden formwork recovery and recycling system includes a screening part, a feeding part, and a processing part. The screening part includes a wood board unit, a wood strip unit and a wood block unit disposed inside a frame unit. The feeding part includes three conveying units inclined and disposed within an installation unit, and ends of the three conveying units are respectively aligned with the wood board unit, wood strip unit and wood block unit. The processing part includes a compaction unit, a moving unit, a discharging unit, a gauze unit, and a glue brushing unit sequentially disposed on a bracket unit. Wood materials are classified into three types and the three-layer conveying mechanism effectively controls an addition order of the wood material, significantly improving an utilization rate of the wood materials and ensuring a uniform density distribution of the compacted wood materials.