Abstract: Provided in the present application are a method and apparatus for determining the capacity of an energy storage apparatus for a new energy support machine, and a new energy support machine. The method includes: determining the duration of a rotor of a doubly-fed induction generator participating in adjustment; determining a slip ratio of the rotor of the doubly-fed induction generator when same participates in the adjustment; and determining the capacity of an energy storage apparatus according to an active power of a stator of the doubly-fed induction generator participating in the adjustment, the slip ratio of the rotor when same participates in the adjustment, and the duration of the rotor participating in the adjustment. The method is intuitive and has a small calculation amount, and a determined capacity value is larger and more reliable, such that the engineering practicability is higher.

Abstract: A speaker and a rollable electronic device include a housing defining an accommodating cavity, a speaker structure arranged in the accommodating cavity, and an inverter tube arranged on the housing. A first sound outlet hole and a communicating hole are respectively in a first side wall and a second side wall of the accommodating cavity. The first side wall is connected to the second side wall. The inverter tube is connected to and extended along the second side wall. The inverter tube includes at least one inverter channel communicated with the accommodating cavity and a second sound outlet hole communicated with the at least one inverter channel. The inverter tube is retractable and stretchable along the second side wall to make the inverter tube be in a blocked state or be in a communicating state. By stretching or compressing the inverter tube, good sound experience is realized.

Abstract: Described herein are method, system, and apparatus for correlating medical images with a medical report. An apparatus may obtain a plurality of medical images of a patient and a medical report for the patient, wherein the plurality of medical images may be associated with respective sequence labels and respective view labels, and wherein the medical report may include multiple diagnoses. The apparatus may determine and fuse respective image-type embeddings and image-content embeddings for the plurality of medical images into a combined image feature representation. The apparatus may further encode the features of the medical report into a text feature representation and calculate a similarity score based on the image feature representation and the text feature representation to indicate a correlation between the medical images and the medical report.

Abstract: An application of a pyrido[1,2-a]pyrimidinone analog or a pharmaceutically acceptable salt thereof in the preparation of a drug, the drug being used to treat and/or prevent one or more among PIK3CA-mutated breast cancer, PIK3CA-mutated ovarian cancer, PIK3CA-mutated endometrial cancer, PIK3CA-mutated cervical cancer, and PIK3CA-mutated bladder cancer.

Abstract: An example of a method includes providing a substrate with an exposed surface comprising a first chemical group, wherein the providing optionally comprises modifying the exposed surface of the substrate to incorporate the first chemical group; reacting the first chemical group with a first reactive group of a functionalized polymer molecule to form a functionalized polymer coating layer covalently bound to the exposed surface of the substrate; grafting a primer to the functionalized polymer coating layer by reacting the primer with a second reactive group of the functionalized polymer coating layer; and forming a water-soluble protective coating on the primer and the functionalized polymer coating layer. Examples of flow cells incorporating examples of the water-soluble protective coating are also disclosed herein.

Abstract: Provided is a die system, and a parameter debugging method and device. The method includes: acquiring measured values of a plurality of key parameters of the die system; determining deviations between the measured values and theoretical values of the plurality of key parameters; determining, according to the deviations of the plurality of key parameters, correlational impact values of the plurality of key parameters on stable production; determining, according to the correlational impact values, whether to trigger parameter debugging on the die system; selecting, in response to a determination to trigger parameter debugging on the die system, a target key parameter from the plurality of key parameters; and performing parameter debugging on the target key parameter.

Abstract: System and method for LLM unlearning via loss adjustments are disclosed. The method includes accessing forget data samples from one or more datasets, associating a template response for each forget data sample via implementation of one or more LLMs, and training a target LLM using a forget data only loss adjustment (FLAT) function to generate an unlearned LLM, including implementing a loss adjustment to maximize a divergence for between an available template answer and a forget answer only with respect to forget data samples.

Abstract: Provided is a method for preparing an electrothermal heating sheet from carbon fiber braided fabric scraps, including: cutting clumps of disordered carbon fiber braided fabric scraps into chopped carbon fibers; washing the chopped carbon fibers by separately using acetone and deionized water, and drying; preparing a corresponding dispersion; adding the chopped carbon fibers to the dispersion, and fully dispersing; performing vacuum filtration by using a double-layer metal screen, and drying to obtain a chopped carbon fiber felt; cutting the chopped carbon fiber felt, sticking electrodes to two ends of the chopped carbon fiber felt, and covering thermoplastic polyurethane (TPU) sheets on front and back surfaces of the chopped carbon fiber felt to form a heating sheet product having electrothermal performance and electromagnetic shielding performance.

Abstract: Certain embodiments of the present application relate to an unmanned aerial vehicle, comprising a chassis, a control system mounted to the chassis, at least one rotor operable to generate lift under control of the control system, a spool having a wound portion of a line wound thereon, the line including a free portion, an attachment device secured to the free portion of the line and configured to facilitate attachment of a parcel to the line, a motor operable to rotate the spool under control of the control system such that the control system is operable to raise the attachment device to a raised position and lower the attachment device from the raised position, and a spring positioned between the chassis and the attachment device such that the spring is engaged between the chassis and the attachment device when the attachment device is in the raised position.

Abstract: Disclosed is an intelligent sports knee brace including a base and a strap. The base is connected to the strap. The strap connects a knee brace to a target part of a target object through a connection structure, and an inner surface of the base facing the target part is provided with an airbag that contacts the target part. A cavity is provided inside the base. An inflation mechanism, a detection mechanism, a battery assembly, an early warning assembly, and a main control board are arranged in the cavity. The detection mechanism is configured to detect air pressure data of the airbag and motion status data of the target object, the inflation mechanism is configured to inflate or deflate the airbag, the early warning assembly is configured to provide a motion warning and an emergency rescue reminder, and the main control board is electrically connected to the inflation mechanism, the detection mechanism, the battery assembly, and the early warning assembly.

Abstract: Provided is a method for trimming and punching a thermoformed part and a trimming and punching die. The method includes the following steps: S1: in a hot stamping die, subjecting a workpiece to hot stamping, initial quenching, and temperature preservation; S2: when a temperature of the workpiece is cooled to a preset temperature, taking out the workpiece and moving the workpiece into a trimming and punching die; S3: determining a to-be-trimmed position and a to-be-punched position, and heating the to-be-trimmed position and the to-be-punched position; S4: after a set condition is reached, trimming and punching the workpiece; and S5: subjecting the workpiece to secondary quenching, so that a hot stamped part can be rapidly punched and trimmed. By accurately controlling cooling rate of the workpiece during machining, temperatures of the workpiece at various positions can be cooled, which helps prevent deformation or cracks and ensures performance of the workpiece.

Abstract: Methods and systems for curating high-quality data samples to enhance Large Language Model (LLM) performance are disclosed. An input prompt corresponding to data samples of one or more datasets related to an enterprise is generated. Based on the input prompt, initial scores for the data samples are generated via implementation of one or more LLMs. Upon generating the input prompt, score curation is performed to correct score errors and to generate curated scores for the data samples. Further, diversity of the data samples is measured to generate long-tail scores for the data samples. The curated scores and the long-tail scores are utilized to determine the high-quality data samples from the data samples. The high-quality data samples are implemented to fine-tune a target LLM.

Abstract: A range-extended numerical controlled link mechanism electric loader comprising a vehicle-loaded rechargeable energy storage device, a vehicle-loaded auxiliary power supply device, a motor, a reducer, a numerical controlled link lever loading mechanism and a rack. The vehicle-loaded rechargeable energy storage device comprises at least a battery, an electric control module, and at least one charging port; the vehicle-loaded auxiliary power supply device comprises an engine and a generator; by cooperation of the vehicle-loaded rechargeable energy storage device and the vehicle-loaded auxiliary power supply device the loader can have five different working modes: low speed mode, high speed mode, hybrid low speed mode, hybrid high speed mode and energy recycling mode, by the different working modes, the wheels can be driven to rotate or the drive levers can be driven to conduct loading work.

Abstract: Systems and methods for generating Generative Artificial Intelligence (Gen AI) responses using context based hierarchical ontological representations are disclosed. In an aspect, input prompt data corresponding to an enterprise is received. A query embedding vector representing the received input prompt data is then generated. Further, at least one relevant node corresponding to the input prompt data is generated. Furthermore, a subset of nodes are selected based on the relevant node. Also, contextual data is generated by aggregating textual content from the selected subset of nodes. An ontological representation corresponding to the input prompt data is generated based on the generated contextual data. A Gen AI response to the input prompt data is then generated using at least one Gen AI model. The generated Gen AI response is then validated. Also, the Gen AI model and the ontological representation are fine-tuned based on results of the validation.

Abstract: A class of compounds that have a pyrazine structure, specifically disclosed is a compound represented by formula, isomers or pharmaceutically acceptable salts thereof, and an application thereof in the preparation of CRAC inhibitors.

Abstract: A tetrahydropyrrolocyclic compound as represented by formula (I), a pharmaceutically acceptable salt thereof, and an application thereof in preparation of a drug for treating a disease related to a selective orexin-2 (OX-2) receptor antagonist, wherein the related disease is selected from insomnia and depression.

Abstract: Embodiments of the present disclosure include compositions and methods for performing in situ hybridization reactions. In particular, the present disclosure provides RNA blocking molecules that enhance detection of a target RNA molecule (e.g., an mRNA molecule, a microRNA (miRNA) molecule, a small non-coding RNA (sncRNA) molecule, a PIWI-interacting RNA (piRNA) molecule, a small interfering RNA (siRNA) molecule, and/or an anti-sense oligo (ASO) molecule) by reducing binding of a target probe to a non-target RNA molecule in a sample.

Abstract: According to the various aspects, a method is provided for dicing a semiconductor panel having a glass core with topside build-up (BU) layers, backside BU layers, and interconnects. In an aspect, a hard mask is deposited on the semiconductor panel and patterned to form openings for a plurality of cut-streets. In an aspect, the dicing of the semiconductor panel includes using plasma dicing steps to form cut-streets through the topside BU layers and the backside BU layers, and using a mechanical sawing step or plasma dicing step to cut through the glass core. In another aspect, the dicing of the semiconductor panel further includes using an acid rinse to remove metal salts when cutting through the glass core during the plasma dicing step. In another aspect, a singulated die may have a first BU sidewall and a second BU sidewall having a morphology that includes semi-sphere fillers.

Abstract: This application provides a rectifier, an inverter, and a wireless charging device. The rectifier includes a signal conversion unit and a switchable capacitor unit that are mutually coupled. The switchable capacitor unit is configured to switch a rectification mode of the rectifier. The rectification mode may include but is not limited to any one of the following: a voltage multiplier rectification mode or a full-bridge rectification mode. Therefore, a variable dynamic range of an output voltage of the rectifier provided in embodiments of this application is large.

Abstract: The techniques disclosed herein provide a system for automatically organizing content captures of user activity into collections based on a shared topic. Due to the significant portion of daily life that occurs via personal computing devices, there is increasing need for helpful user experiences to enhance productivity and engagement. Namely, those directed to assisting a user in managing and recalling their past activity. As such, the presented system retrieves content captures depicting a moment of interest. Accordingly, the system performs a visual analysis of each content capture to determine a semantic context and assign one or more topics. This is accomplished by utilizing a pair of machine learning models to identify and extract relevant text information. The extracted text information is then analyzed by a topic classification model to assign the one or more topics. The system further includes graphical user interfaces for viewing and managing collections of content captures.