Abstract: A sequence conversion method includes receiving a source sequence, converting the source sequence into a source vector representation sequence, obtaining at least two candidate target sequences and a translation probability value of each of the at least two candidate target sequences according to the source vector representation sequence, adjusting the translation probability value of each candidate target sequence, selecting an output target sequence from the at least two candidate target sequences according to an adjusted translation probability value of each candidate target sequence, and outputting the output target sequence. Hence, loyalty of a target sequence to a source sequence can be improved during sequence conversion.

Abstract: An air spring having an airsleeve, wherein at least one layer of the airsleeve includes the vulcanization product of a composition that includes at least one vuicanizable polymer, a curative, at least one halogenated hydrocarbon wax, and expandable graphite.

Abstract: One example method includes obtaining L1 first decoding paths of an (i?1)th group of to-be-decoded bits, where i is an integer, received data corresponds to P groups of to-be-decoded bits, and 1<i?P, determining at least one second decoding path corresponding to each first decoding path, where a quantity of second decoding paths corresponding to each first decoding path is less than 2n, and where n is a quantity of information bits included in an ith group of to-be-decoded bits, and determining at least one reserved decoding path of the ith group of to-be-decoded bits in second decoding paths corresponding to the L1 first decoding paths. The at least one reserved decoding path includes a decoding result of the ith group of to-be-decoded bits.

Abstract: A method of location determination with a WiFi transceiver and an AI model includes jointly training, based on various losses: a feature extractor, a location classifier, and a domain classifier. The domain classifier may include a first domain classifier and a second domain classifier. The losses used for training tend to cause feature data from the feature extractor to cluster even if a physical object in an environment has moved after training is completed. Then, the location classifier is able to accurately estimate the position of, for example, a person in a room, even if a door or window has changed from open to close or close to open between the time of training and the time of estimating the person's position.

Abstract: Described herein are techniques and systems for retrieval of content using link-based searches. Embodiments herein include receiving, by a server, a request including a hyperlink directed to a source content. The server may determine whether a database associated with the server comprises the hyperlink. The database stores information of multiple entities each corresponding to one or more hyperlinks. In response to a determination that the database includes the hyperlink, the server may identify an entity corresponding to the hyperlink and provide information of the entity to the computing device. In some embodiments, in response to a determination that the database does not include the hyperlink, the server may identify related web content and return results to user.

Abstract: Disclosed are an electric wheel assembly applicable for a parallelly mounted dual-tire wheel, an axle, and a vehicle. The electric wheel assembly includes a wheel and support axle assembly, an internal-rotor hub motor, a planetary gear reducer and a brake system. A planetary carrier is connected to a hub. The hub is connected to a spoke and a rim. The wheel support axle passes through a rotor sleeve and is provided with a hub bearing to support the hub and a whole wheel. A sun gear disposed at an end of the rotor sleeve of the internal-rotor hub motor is a power input of the planetary gear speed reducer, and the planetary carrier is a power output. A planetary gear of the planetary gear reducer is a tower-shaped gear and realizes a relatively high transmission ratio under a condition that an axial length is relatively small.

Abstract: Example successive cancellation list (SCL) parallel decoding methods and apparatus are described. One example method includes obtaining L1 first decoding paths of an (i?1)th group of to-be-decoded bits after received data corresponds to P groups of to-be-decoded bits, where i is an integer, P is an integer greater than 1, 1<i?P, and L1 is a positive integer. L3 third decoding paths is determined for each first decoding path, where a quantity of information bits in an ith group of to-be-decoded bits is n, n is a positive integer greater than or equal to 1, L3 is a positive integer, and L3<2n. At least one reserved decoding path of the ith group of to-be-decoded bits is determined from L1×L3 third decoding paths, where the at least one reserved decoding path includes a decoding result of the ith group of to-be-decoded bits.

Abstract: A method and system for generating permissions policies and permission boundary policies are described. The system receives a first request from a central administrator to create a delegated administrator, the first request specifying with one or more access permissions. The system generates a permission boundary policy that specifies the one or more access permissions and a first permissions policy that grants permissions to the delegated administrator to at least one of create an IAM principal with the permission boundary policy or attach a second permissions policy to the IAM principal. An effective permission given to the IAM principal is an intersection of access permissions specified in the first permissions policy and the one or more access permissions in the permission boundary policy. The system attaches the first permissions policy and the permission boundary policy to the delegated administrator.

Abstract: A flatness sensing device includes a pressure sensing unit and a controller. The pressure sensing unit includes a plate portion, and a pressure sensor fixed and exposed at an edge of the plate portion. A sensing surface of the pressure sensor is flush with the bottom surface of the plate portion. The pressure sensor senses a pressure or lack of pressure of an object to be tested when it touches the object to be tested. The controller receives the pressure value output from the pressure sensor and determines a flatness of the object to be tested, the controller can issue an alarm on finding non-flatness and cause the degree of flatness to be displayed to a user.

Abstract: An information displaying method and a terminal are provided. The method includes: obtaining audio data to be played in a chronological order; determining, based on attribute information at any moment of a sound represented by the audio data, a shape of a graph corresponding to the any moment, where the graph corresponding to the any moment including a closed curve with a bump, and a maximum distance in distances from points on the bump to a center of the graph is positively correlated to a value indicated by the attribute information at the any moment; and displaying the graph corresponding to the any moment. The bump in the graph changes with the value indicated by the attribute information of the sound, and such graph is presented to a user, to enhance perception of the user on the attribute information of the audio data and improve user experience.

Abstract: An object recommendation method and apparatus, a storage medium and a terminal device are provided. The method includes: obtaining historical behavior data and historical feedback data of a user; determining a questioning keyword based on the historical behavior data and the historical feedback data; performing a question-and-answer interaction based on the questioning keyword to obtain feedback data; determining a target recommendation object based on the feedback data; and outputting the target recommendation object, so as to implement recommendation. The provided technical solution diversifies the dimensions in prediction of the interest tendency of the user by using the question-and-answer interaction, improves the accuracy and flexibility in recognizing the objects of interest of the user, thereby improving the accuracy and flexibility of recommendation.

Abstract: A method for generating a digital identity is provided, including: extracting a first preset number of short tandem repeat STRs and relevant information of each STR from whole genome data; generating a single STR digital code corresponding to each STR according to the relevant information of each STR, to obtain a plurality of single STR digital codes; performing sequence transformation on each single STR digital code with a preset rule, and generating a target STR digital code according to the single STR digital code after the sequence transformation; generating summary information of the target STR digital code, and determining the summary information as summary information of the STR to which the target STR digital code belongs; and determining the summary information of the STR as the generated digital identity.

Abstract: A high-speed connector for an automobile includes a cable connector, a board connector connected to the cable connector, and sliding blocks provided on two sides of the cable connector. The sliding blocks are separate or are integrated. The cable connector includes a housing, a contact body and a cable. The contact body is installed in the housing. The contact body is in contact with the board connector to achieve an electric connection. The cable is connected to the contact body and protrudes out of the housing. The housing includes a main body portion and side snap-fit plates provided on two sides of the main body portion. A side snap-fit groove is provided on the side snap-fit plate. Side snap-fit clasps are provided on two sides of the board terminal connector. The sliding blocks are installed on the side snap-fit plates and slide upwards and downwards along the side snap-fit plates.

Abstract: A location-aware electronic device is provided. The electronic device trains feature extraction layers, reconstruction layers, and classification layers. The training may be based on a reconstruction loss and/or a clustering loss. The electronic device processes a fingerprint to obtain an augmented fingerprint using randomization based on statistics of the fingerprint. The feature extraction layers provide feature data to both the reconstruction layers and the classification layers. The classification layers operate on the codes to obtain an estimated location label. An application processor operates on the estimated location label to provide a location-aware application result to a person.

Abstract: An ultra high speed signal cable connector includes an insulating housing, a PCB fixedly mounted inside the insulating housing, and at least one row of cables without grounding wires welded to the PCB. The end portion of a connecting finger of the PCB extends out of the insulating housing, and the PCB is provided with a signal bonding pad and a grounding wire bonding pad. Each cable without grounding wires includes a plurality of electric wires, a conductive jacket for wrapping the plurality of electric wires together, and an insulating jacket wrapped on the outer layer of the conductive jacket. Each electric wire includes a core wire and an insulating layer wrapped on the outer layer of the core wire. One ends of the core wires are welded on the signal bonding pad; a section of insulating jacket is stripped from the foremost end of each cable without grounding wires to expose the conductive jacket. The conductive jackets are electrically and fixedly connected to the grounding wire bonding pad.

Abstract: A MEMS micromirror includes a mirror surface driving structure which is positioned on a substrate and includes two L-shaped structures in head-to-tail arrangement. Each L-shaped structure includes a second torsion beam, an L-shaped transverse plate and a second comb-shaped structure. The first driving electrode is provided on the substrate at a position under a head end of the L-shaped transverse plate, the head end of the L-shaped transverse plate is rotatable with support of the second torsion beam, and a tail end of the L-shaped transverse plate is connected with the second comb-shaped structure. The micromirror surface layer is disposed above the mirror surface driving structure, the first torsion beam is fixed by the substrate and supports two sides of the micromirror surface layer, and two sides, corresponding to the second comb-shaped structures, of the micromirror surface layer are provided with first comb-shaped structures, respectively.

Abstract: Described herein are techniques and systems for retrieval of content using link-based searches. Embodiments herein include receiving, by a server, a request including a hyperlink directed to a source content. The server may determine whether a database associated with the server comprises the hyperlink. The database stores information of multiple entities each corresponding to one or more hyperlinks. In response to a determination that the database includes the hyperlink, the server may identify an entity corresponding to the hyperlink and provide information of the entity to the computing device. In some embodiments, in response to a determination that the database does not include the hyperlink, the server may identify related web content and return results to user.

Abstract: In one example, a computing device includes one or more user input detection components, and one or more processors configured to receive an indication of a first user input detected by the one or more user input detection components, responsive to receiving the indication of the first user input, adjust a level of an attention buffer at a defined rate; responsive to determining that the level of the attention buffer satisfies a first threshold, prevent further interaction with a user interface of the computing device, responsive to determining that an indication of a second user input has not been received within a time period, adjust a level of the attention buffer, and responsive to determining that the level of the attention buffer satisfies a second threshold, allow further interaction with the user interface.