Abstract: The present disclosure provides a memory system packaging structure and fabrication methods. The memory system packaging structure includes memory modules, a memory controller, a redistribution layer electrically connected to the memory controller, a plastic encapsulation layer encapsulating the memory modules and the memory controller, and one or more connecting pillars extending in the vertical direction and configured for providing electric power to the memory modules. Each memory module includes memory dies stacked in a vertical direction. Each connecting pillar includes a first portion being in physical contact with one of the memory dies and a second portion being in physical contact with the redistribution layer.

Abstract: A data processing method is performed by a first device and includes generating first data feature information based on first data using a first model, the first data feature information being configured for characterizing a data feature of the first data; adding noise information to the first data feature information, to obtain second data feature information; and transmitting the second data feature information to a second device, to instruct the second device to generate a first data processing result based on the second data feature information using a second model, the first model and the second model being configured for forming a data processing model, and the first data processing result being configured for characterizing a data processing result obtained by processing the first data using the data processing model.

Abstract: A driving circuit is configured to drive a light-emitting unit to emit light, and the driving circuit includes: a voltage regulation circuit, configured to regulate a driving voltage of the light-emitting unit according to a voltage regulation signal; and a processing chip, configured to provide the voltage regulation signal to the voltage regulation circuit to reduce a difference between a reference current and a driving current of the light-emitting unit under a turned-on state.

Abstract: A region recognition method includes obtaining a scene image of a scene to-be-recognized; determining a specified boundary curve of the scene image; generating a contour map corresponding to the scene image according to the specified boundary curve; and determining a target region in the scene according to the contour map.

Abstract: A model training method and apparatus for federated learning, a device and a storage medium are provided, which belong to the technical field of machining learning. The method includes: generating an ith scalar operator based on a (t?1)th round of training data and a tth round of training data (201); transmitting an ith fusion operator to a next node device based on the ith scalar operator (202); determining an ith second-order gradient descent direction of an ith sub-model based on an acquired second-order gradient scalar, an ith model parameter and an ith first-order gradient; and updating the ith sub-model based on the ith second-order gradient descent direction to obtain a model parameter of the ith sub-model during a (t+1)th round of iterative training.

Abstract: A video frame extraction method includes uploading a video through an online access link and locally caching the video to a local database; generating a local access link of the video according to the online access link that includes a local cache identifier and a local video identifier; in response to a video frame extraction request for the video, obtaining a playback time of target video frame of the video for extracting and triggering an access request and carrying the playback time of target video frame; forwarding the access request to the local database in response to identifying the local cache identifier from the local access link, to enable the local database to return a locally cached video that matches based on the local video identifier; and obtaining the locally cached video returned by the local database and extracting a video frame matching the playback time of target video frame.

Abstract: The present application provides an integrated package device, a method of fabricating an integrated package device, and a memory system. The integrated package device may include at least one package module. Each package module may include a first sub-package module with a first electronic devices. Each package module may include a second sub-package module including a second electronic devices. Each package module may include a first re-distribution layer connected with first pads. Each package module may include a second re-distribution layer connected with second pads.

Abstract: Provided are anti-CLDN18 antibodies or antigen-binding fragments thereof, isolated polynucleotides encoding the same, pharmaceutical compositions comprising the same, and the uses thereof.

Abstract: A method includes obtaining a training dataset including a plurality of training utterances each paired with a corresponding transcription. For each respective training utterance the operations include generating, using a large language model (LLM), a corresponding speech recognition result of the respective training utterance; determining a corresponding ground-truth histogram based on the corresponding transcription, determining a corresponding prediction histogram based on the corresponding speech recognition result; and determining a corresponding word-level hallucination loss based on the corresponding ground-truth histogram and the corresponding prediction histogram. The corresponding ground-truth histogram represents a number of occurrences of each token in the corresponding transcription, while the corresponding prediction histogram represents a number of occurrences of each token in the corresponding speech recognition result.

Abstract: The present disclosure relates to a tape removing device and a tape removing method. The present disclosure provides a tape removing device comprising: a working table; a suction nozzle; a workpiece for tape removing, the workpiece for tape removing including a support adhesive layer and at least one workpiece bonded to a side of the support adhesive layer, in which one side of the support adhesive layer provided with the at least one workpiece is facing the working table; and a guiding roll in contact with the other side of the support adhesive layer, wherein the suction nozzle fixes the workpiece for tape removing on the working table. Therefore, it is possible to avoid stripping off the support adhesive layer from the workpiece at a right angle, thereby reducing damages to the workpiece, which in turn can reduce the risk of invisible product cracks and improve product reliability.

Abstract: This application discloses a photovoltaic solar cell, a manufacturing method thereof, and a photovoltaic module. The photovoltaic solar cell structure includes: a back contact solar cell, a soldering portion, an insulating material, and a conductive material. Each of the soldering portion is at least located on a corresponding busbar included in the back contact solar cell. The insulating material covers at least a target part of each finger included in the back contact solar cell. The target part is a part in which a distance between each of the finger and a busbar having a polarity opposite to that of the finger is less than a preset distance. The conductive material is formed on each of the soldering portion, and a height of a top of the conductive material is greater than or equal to a height of a top of the insulating material.

Abstract: The present disclosure relates to a distributed training method, which includes: a terminal sends a first message to a base station, the terminal being a distributed node for distributed learning, the base station being a central node for distributed learning, and the first message comprising training information for a model deployed by the terminal; and the terminal receives a second message sent by the base station, the second message comprising at least one of a parameter update instruction and a convergence condition update instruction of the model, or comprising a training stop instruction, the first message and the second message being control channel signalings and being scrambled by using a scrambling code sequence, and the scrambling code sequence being determined according to at least one of a model function, a model identifier, a training mode and a training group to which the terminal belongs.

Abstract: A crankshaft assembly includes a crankshaft body extending along an axis of rotation. The crankshaft body includes bearing journals that are mutually coaxial with the axis of rotation and spaced from each other along a length of the crankshaft body and crankpins that are spaced from each other along the length of the crankshaft body and axially offset from the axis of rotation. The crankshaft body also includes crank webs projecting radially from the axis of rotation and interconnecting the bearing journals and the crankpins and a flange extending from a distal end of the crankshaft body having a conical gear engaging surface.

Abstract: The present disclosure provides anti-CLDN18.2 antibodies or antigen-binding fragments thereof, pharmaceutical com-position comprising the same and the uses thereof (e.g., diagnostic use).

Abstract: An interaction method for a mobile robot comprises: acquiring map data information of a space in which a mobile robot is located, and acquiring real-time environment perception data collected by an environment perception sensor, the real-time environment perception data comprising real-time obstacle information, and real-time indication information for indicating a road condition around the mobile robot; acquiring target traveling path information of the mobile robot on the basis of the real-time obstacle information and the map data information, and determining a ground projection region according to the target traveling path information and the real-time indication information; acquiring a pattern to be projected, and determining a projection parameter corresponding to said pattern, wherein said pattern is used for indicating a traveling intention of the mobile robot; and controlling a projection apparatus according to the projection parameter, so as to project said pattern onto the ground projection region

Abstract: The present disclosure pertains to non-porous composite particles that are non-porous, polymer-based particles. In various embodiments, a non-porous polymer core is surface modified. In various embodiments, a non-porous hybrid organic-inorganic material is in contact with the modified surface of the core, and a bonding material is in contact with the non-porous hybrid organic-inorganic material. The present disclosure pertains to chromatographic separation devices that comprise such non-porous composite particles.

Abstract: This invention discloses a double-liquid grouting slurry, its technology and application for super large diameter underwater shield engineering under high water pressure condition. The materials of slurry I are: 35-45 parts of cement clinker; 15-25 parts of slag; 24-35 parts of fly ash; 15-25 parts of steel slag; 5-15 parts of bentonite; 4-10 parts of limestone tailing; 0.3-2.0 parts of water reducing agent; 0.5-2.5 parts of cellulose. The materials of slurry II are: 0.2-3.8 parts of short-cut fiber; 96-99 parts of sodium silicate solution; 0.8-4.8 parts of viscous polymers. This invention generates the double-liquid slurry preparation process including crushing-screening-milling-group mixing-grouped mixing at different speeds, the volume ratio of slurry I and II is 1:1-10:1 during grouting, and the slurry is injected into the shield void through the six-point position technology at the shield tail and 3+2+1 segment splicing synchronous grouting techniques.

Abstract: A construction process for partially bonded prestressed uplift-resistant glass fiber anchor piles avoiding stress concentration includes the following steps: construction site management, determining anchor pile locations, drilling a borehole using a spiral drilling rig, lowering a pre-fabricated glass fiber rod cage into the borehole after inspection, performing grouting, making adjustments, and repeating the above operations. The construction process adopts a layout that avoids stress concentration, demonstrating its superiority in strata with developed horizontal bedding. Simultaneously, it uses glass fiber material with good corrosion resistance, meeting the requirements for uplift-resistant anchors under corrosive conditions. It employs a partially bonded prestressed design, changing the mechanical characteristic of traditional uplift-resistant anchors from becoming looser when pulled to a wedge-like characteristic of becoming tighter when pulled.

Abstract: The present disclosure provides a method and apparatus for adjusting timing, a computer-readable storage medium and an electronic device. The method for adjusting timing includes: in response to a data stream outputted by a transmitting end of a display port being received, determining a first line period according to the data stream, and the first line period is a line period of the data stream under an output clock, and the output clock is a clock domain of a receiving end of the display port; processing the first line period at least using a Kalman filtering algorithm to obtain a second line period, and the second line period is a line period of the data stream under a source clock, and the source clock is a clock domain of the transmitting end; and adjusting the output clock according to the second line period.

Abstract: The present disclosure provides an aza[3.2.1]bridged ring substituted fluorescent dye and a preparation method and use thereof. According to the fluorescent dye, by adopting an aza[3.2.1]bridged ring as a substituent of the fluorescent dye, not only are the brightness and photostability of the fluorescent dye obviously improved, but also the phenomenon of spectral blue shift caused by irradiation of the dye for a long time is completely avoided. Meanwhile, the water solubility of the fluorescent dye can be effectively improved, and good membrane permeability of the fluorescent dye is maintained. The present disclosure further provides embodiments of functionalized derivatives of the dye in single molecule imaging, confocal imaging and super-resolution imaging of biological samples.