Abstract: Provided are a large language model (LLM) training method and apparatus, a device, and a storage medium. The method includes: acquiring a corpus data training set to be trained, and selecting corpus data from the corpus data training set; predicting the corpus data by using an LLM, obtaining a prediction result, calculating a current loss value based on the prediction result, saving the current loss value, and calculating a current preset loss value based on the saved current loss value; and determining whether the current loss value is less than the current preset loss value; if yes, skipping updating model parameters of the LLM; if not, updating the model parameters of the LLM and the current preset loss value based on the current loss value, re-selecting corpus data from the corpus data training set, and continuing to update the model parameters of the LLM until a trained LLM is obtained.

Abstract: The present disclosure provides techniques for automated relationship discovery. One example method includes receiving input data indicative of a plurality of heterogeneous data entries, wherein the plurality of heterogeneous data entries indicate a type of subjects, generating, for each heterogeneous data entry in the plurality of heterogeneous data entries, an individual graph, wherein nodes of the plurality of individual graphs represent the type of subjects, and wherein the nodes indicate a plurality of attributes, generating a unified graph based on the plurality of individual graphs, receiving an indication associated with a particular node in the unified graph, identifying a set of neighboring nodes associated with particular node, determining an action to be applied to the particular node, and applying the action to each node in the set of neighboring nodes.

Abstract: A use of a soybean C2H2 type zinc finger protein transcription factor GmZFP7 with the amino acid sequence as shown in SEQ ID NO: 2, or a gene GmZFP7 with a gene accession number of Gene ID: 100792169 in regulating isoflavones is provided. Also provided is a method for regulating isoflavones, which increases or reduces a total isoflavone content in a plant by way of overexpression, silencing, knockdown or knockout of the transcription factor gene GmZFP7.

Abstract: A method for repairing a virtual machine in a cloud environment is provided. A server device receives a first request that is transmitted by a console, the first request includes a virtual machine identifier of a target virtual machine and requests a repair of the target virtual machine. An image database is queried for a target image identifier of a pre-made Live CD image corresponding to the virtual machine identifier. A target cloud block storage is created for storing the Live CD image. The target cloud block storage is mounted to a host machine that hosts the target virtual machine to cause the host machine to download the Live CD image from an image repository to store to the target cloud block storage, mount the target cloud block storage to the target virtual machine, and boot the target virtual machine via the target cloud block storage.

Abstract: Provided herein are methods for determining whether a disorder is susceptible to treatment with a plasma kallikrein inhibitor and methods for identifying a subject as a candidate for treatment with a plasma kallikrein inhibitor. Provided herein are methods of identifying a subject as having or being at risk for a disorder and methods of treating a disorder in a subject. Also provided herein are methods of evaluating effectiveness of a treatment in a subject.

Abstract: A system, device, system-on-a-chip, and method of automatically correcting an object not found error using image recognition are described. The method includes running a test script for testing and analysis of a web page as rendered by a web browser. The method further includes, responsive to detecting the object not found error, automatically locating a missing object associated with the object not found error. One method of locating a missing object includes using image recognition. The method also includes updating the test script with a located object. The method may also include replaying the test script.

Abstract: A sorting vehicle is provided that includes a wire harness structure and at least two vehicle bodies. The at least two vehicle bodies are pivotally connected to one another in sequence, the wire harness structure is connected to the two adjacent vehicle bodies, and the wire harness structure includes an elastic deformation portion located at a junction of the two adjacent vehicle bodies.

Abstract: A portable temperature-regulating device wearable as a backpack or chest-mounted unit comprises a bag body and at least two shoulder straps connected to an upper portion of the bag body for securing the device to a user's shoulders or chest. The bag body defines an air inlet and a main cavity air duct along which a fan mounting portion and a temperature-regulating element mounting portion are disposed. When a fan and a temperature-regulating element are installed, airflow entering the main cavity air duct through the air inlet is conditioned to form a temperature-regulated airflow. Each shoulder strap includes a branch cavity air duct having at least one air outlet hole, the branch cavity air duct being in fluid communication with the main cavity air duct such that the temperature-regulated airflow enters the branch cavity air duct and is discharged from the air outlet hole toward temperature-sensitive areas of the user.

Abstract: The photovoltaic cell includes a silicon substrate, a first passivation layer, a second passivation layer, at least one silicon oxynitride layer, and at least one silicon nitride layer. The second passivation layer includes a first silicon oxide layer and at least one aluminum oxide layer, and a thickness of the at least one aluminum oxide layer is in a range of 4 nm to 20 nm. The number of silicon atoms is greater than the number of oxygen atoms in the at least one silicon oxynitride layer and the number of oxygen atoms is greater than the number of nitrogen atoms in the at least one silicon oxynitride layer. The first silicon oxide layer is disposed between the substrate and the at least one aluminum oxide layer, and a thickness of the first silicon oxide layer is in a range of 0.1 nm to 5 nm.

Abstract: The invention uses gel layers to cover and/or separate one or more aqueous reaction mixtures in a reaction vessel. A gel layer can protect reagents or reaction products from contacting the atmosphere. A gel layer can also separate multiple aqueous reaction mixtures so that they do not contact one another until the gel layer is broken. An oil layer can be inserted immediately underneath a gel layer. Gel layers can be penetrated by insertion of a tip from a micropipette to deliver a solution, such as a sample, below a gel layer. Micropipette tips or inert rods can also be used to disrupt gel layers resulting in mixing of aqueous reaction mixtures above and below a layer. Gel layers can also be broken by selective application of heat or movement of magnetic beads. Multiple gel layers separating multiple aqueous reagent mixtures can be broken in a defined sequence for an ordered series of bioprocessing steps.

Abstract: Systems and methods are provided for calculating, using an electronic processor, an average environmental brightness and determining a current pulse width modulation (“PWM”) output level provided to the light source. The method also includes determining, using the electronic processor, a target illumination level and a PWM adjustment rate. The PWM adjustment rate is based at least partially on the calculated average environmental brightness. The method also includes adjusting, using the electronic processor, the current PWM output level at the determined PWM adjustment rate to reach the target illumination level, and transmitting the adjusted PWM output level to the light source. The target illumination level is determined as a function of the current PWM output level and an output mode of the light source.

Abstract: The photovoltaic cell includes a silicon substrate, a first passivation layer, a second passivation layer, at least one silicon oxynitride layer, and at least one silicon nitride layer. The second passivation layer includes a first silicon oxide layer and at least one aluminum oxide layer, and a thickness of the at least one aluminum oxide layer is in a range of 4 nm to 20 nm. The number of silicon atoms is greater than the number of oxygen atoms in the at least one silicon oxynitride layer and the number of oxygen atoms is greater than the number of nitrogen atoms in the at least one silicon oxynitride layer. The first silicon oxide layer is disposed between the substrate and the at least one aluminum oxide layer, and a thickness of the first silicon oxide layer is in a range of 0.1 nm to 5 nm.

Abstract: A method for rendering output audio via a wearable audio device is provided. The method includes: (1) generating an aware audio signal based on a playback audio signal and an environmental aware signal corresponding to a feedforward microphone; (2) filtering, via an adaptive filter, the aware audio signal to generate an adapted aware signal; (3) generating an acoustic driver signal based on the adapted aware signal and a feedback audio signal corresponding to a feedback microphone; and (4) rendering, via an acoustic driver, output audio based on the acoustic driver signal. The adaptive filter is configured to adapt based on the aware audio signal and the feedback audio signal.

Abstract: Pyrazole carboxylate ester compounds have a structural formula of formula (I): Q is Q1 or Q3 The pyrazole carboxylate ester compounds has herbicidal activity and can be used for controlling weeds.

Abstract: According to implementations of the subject matter described herein, a solution for neural video coding is provided. According to the solution, estimated motion information of a target frame in a video, reference feature information and a reference reconstructed frame of a reference frame for the target frame are obtained. Context information for the target frame is determined using a context extraction model and based on the estimated motion information, the reference reconstructed frame, and the reference feature information. In a conversion between the target frame and a bitstream of the video, a reconstructed target frame of the target frame is generated using a frame coding model and based on at least the context information. In this way, richer context information can be extracted for coding and thus the coding efficiency can be improved.

Abstract: A topical formulation for treating atopic dermatitis (AD), and a preparation method and use thereof, belonging to the technical field of development of traditional Chinese medicine (TCM) topical formulations. An active ingredient of the topical formulation for treating AD is neoliquiritin with a concentration of 0.5 wt % to 1.5 wt %.

Abstract: The present disclosure provides methods, compositions, kits and systems for nucleic acid amplification. In some embodiments, nucleic acid amplification methods include subjecting the nucleic acid to be amplified to partially denaturing conditions. In some embodiments, nucleic acid amplification methods include amplifying without fully denaturing the nucleic acid that is amplified. In some embodiments, the nucleic acid amplification method employs an enzyme that catalyzes homologous recombination and a polymerase. In some embodiments, methods for nucleic acid amplification can be conducted in a single reaction vessel and/or in a single continuous liquid phase of a reaction mixture, without need for compartmentalization of the reaction mixture or immobilization of reaction components.

Abstract: Disclosed are a sagger for sintering lithium composite transition metal oxide and a preparation method thereof. The sagger includes a substrate layer and a shallow layer on a surface of the substrate layer, and a coating layer. The substrate layer is prepared from the following raw materials: silicon carbide, magnesia-alumina spinel, aluminum oxide-magnesium oxide-yttrium oxide composite fiber, zircon powder and a binding agent; the shallow layer is prepared from the following raw materials: silicon carbide, magnesia-alumina spinel, aluminum oxide-titanium oxide composite fiber, yttrium oxide-zirconium oxide composite fiber and a binding agent; and the coating layer is prepared from the following raw materials: silicon carbide, magnesia-alumina spinel, magnesium oxide, zirconium oxide fiber, lithium composite transition metal oxide powder and a binding agent. The sagger of the present disclosure has properties of good corrosion resistance and a small coefficient of thermal expansion.

Abstract: Provided is an adjustable lighting device that includes a housing, a circuit board provided in the housing, and the circuit board has first circuit terminals and second circuit terminals, light-emitting units, connected to the circuit board, switch terminals arranged on the circuit board for controlling the turning on and turning off of the plurality of light-emitting units, and a sliding rod being slidingly mounted on the housing and contacting with the switch terminals, the second circuit terminals are respectively connected to the light-emitting units, and when the sliding rod contacts and interacts with one of the switch terminals, the switch terminal selectively connects the first circuit terminals to the second circuit terminals, respectively, to control at least some light-emitting units to be turned on or off, to change the turning-on states of the light-emitting units, thereby adjusting lighting parameters of the lighting device.