Abstract: A method for manufacturing an out plate for a refrigerator includes providing a steel plate with a protective film attached to a surface thereof; seating the steel plate on a die and forming one of the plurality of through holes by a vertical movement of a punch; and horizontally moving the steel plate to one of a plurality of set positions to form another one of the plurality of through holes in the steel plate by the vertical movement of the punch, wherein the out plate defines a front surface of a door of the refrigerator and includes a display part, wherein a plurality of through holes are formed therein, and a display assembly comprising a plurality of light emitting diodes (LEDs) that are configured to transmit light through the plurality of through holes configured to display information.

Abstract: Provided is a hydrogenation method of a phthalate-based compound. According to the present invention, after hydrogenation of the phthalate-based compound, a reaction product is stably decompressed to separate and recover unreacted hydrogen, thereby improving stability and economic efficiency of the process.

Abstract: Disclosed is a method of generating data based on input data by using a pre-trained artificial neural network model having an encoder-decoder structure. In particular, according to the present disclosure, a computing device generates new data based on a probability distribution of input data by using a pre-trained artificial neural network model having an encoder-decoder structure, and the pre-trained artificial neural network model having the encoder-decoder structure corresponds to a pre-trained model in which a latent vector layer is included between an encoder layer and a decoder layer of the artificial neural network model.

Abstract: A method for quantization learning by a model quantizer that is operating in a computer system and compressing a transformer model. The method may include generating a student model through quantization of the transformer model, performing a first quantization learning by inserting a self-attention map of a teacher model into a self-attention map of the student model, and performing a second quantization learning using a knowledge distillation method so that the self-attention map of the student model follows the self-attention map of the teacher model.

Abstract: Disclosed is a method for identifying a word corresponding to a target word in text information, which is performed by one or more processors of a computing device. The method may include: determining a target word; determining a threshold for an edit distance associated with the target word; determining a word of which the edit distance from the target word among words included in text information is equal to or less than the threshold; and identifying the word corresponding to the target word based on the determined word.

Abstract: Disclosed are a method, a device, and a program for selecting a main speaker among speakers included in a sound source or a conversation record based on the sound source or the conversation record including conversation contents of at least one speaker and generating a summary based on the main speaker. A method of generating a summary for a sound source, the method being performed by at least one computing device, includes: generating a speak score for at least one speaker based on the sound source; determining a main speaker of the sound source based on a speak score for said at least one speaker; and generating a summary for the sound source in consideration of the determined main speaker.

Abstract: Disclosed is a method of generating data based on input data by using a pre-trained artificial neural network model having an encoder-decoder structure. In particular, according to the present disclosure, a computing device generates new data based on a probability distribution of input data by using a pre-trained artificial neural network model having an encoder-decoder structure, and the pre-trained artificial neural network model having the encoder-decoder structure corresponds to a pre-trained model in which a latent vector layer is included between an encoder layer and a decoder layer of the artificial neural network model.

Abstract: An apparatus and method with neural network operation are provided. A computing apparatus includes one or more processors, storage hardware storing instructions configured to, when executed by the one or more processors, cause the one or more processors to: extract calibration data from training data that is for training a main neural network, based on the calibration data, generate a look up table (LUT) for performing a non-linear function of the main neural network through an auxiliary network corresponding to a layer of the main neural network, and update a parameter of the LUT based on an output of the non-linear function and based on an output of the auxiliary network.

Abstract: Disclosed is a method for speech recognition performed by one or more processors of a computing device. The method includes inputting voice information into an encoder to extract a first feature vector and calculating a first loss function. The method includes inputting the first feature vector extracted from the encoder to a first decoder to perform prediction on the voice information, calculating a second loss function, and extracting a second feature vector. The method includes inputting a second feature vector extracted from the first decoder to a second decoder to perform grapheme-based prediction, and calculating a third loss function. The method includes training at least one of the encoder, the first decoder, or the second decoder based on the first loss function, the second loss function, and the third loss function.

Abstract: Disclosed is a method for generating data, the method is performed by one or more processors of a computing device. The method may include: segmenting text data generated based on speech information into a token unit; generating a first feature vector based on the text data segmented into the token unit, and generating a first label vector corresponding to the generated first feature vector, and generating a second feature vector and a second label vector by performing mix-up for each of the generated first feature vector and the generated first label vector.

Abstract: Disclosed is a method for dialogued summarization with word graphs, which is performed by one or more processors of a computing device. The method may include: generating a word graph based on information on a dialogue which is a summary target; extracting at least one keyword based on the generated word graph; generating a plurality of candidate summary sentences based on the generated word graph; and calculating a score associated with at least one keyword for each of the plurality of candidate summary sentences, and selecting at least one of the plurality of candidate summary sentences based on the calculated score.

Abstract: This invention relates to a preparation method of an ester compound that can reduce energy consumption due to convenient post-treatment process, and simultaneously, can produce products of high purities.

Abstract: This invention relates to a process for hydrogenation of a phthalate based compound. According to the invention, generation of by-products is inhibited during hydrogenation, and thus, catalytic activity is improved and life is prolonged, thereby increasing efficiency and economical feasibility of the industrial process. And, since the hydrogenation product prepared according to the invention has high purity and low acid value, it has excellent qualities as a plasticizer, and thus, can be used for various products.

Abstract: According to an aspect, a method of preparing quantum dots includes a first operation of preparing a quantum dot seed solution; a second operation of growing a quantum dot by continuously injecting a quantum dot cluster solution into the quantum dot seed solution; a third operation of separating the grown quantum dot and dispersing the quantum dot in a solvent; and a fourth operation of further growing the quantum dot by continuously injecting the quantum dot cluster solution into the dispersed quantum dot.

Abstract: A data catalog providing method configured to provide functions related to management and retrieval for data sets stored in a database is provided. The data catalog providing method provides recommendation information for a user by collecting log data of users querying a data set by using a data catalog, and using AI (Artificial Intelligence) recommendation model, based on log data and/or data sets. The AI recommendation model, which is learned based on the collected log data, generates recommendation information by using different recommendation algorithms according to an amount of the accumulated log data.

Abstract: Provided is a flexible substrate. The flexible substrate includes a first film having a first Young's modulus, a second film on the first film and having a second Young's modulus, and a third film between the first film and the second film and having a third Young's modulus. The third Young's modulus is less than each of the first Young's modulus and the second Young's modulus.

Abstract: A removing poison apparatus of a lambda sensor includes: a lambda sensor detecting an oxygen concentration included in an exhaust gas; and a control unit differentiating a heating temperature and a heating time to remove poison of the lambda sensor depending on a lambda signal output from the lambda sensor in an overrun condition of the engine. By heating the lambda sensor depending on the magnitude of the lambda signal output from the lambda sensor, the lambda sensor may be inhibited from being degraded by combusting the material poisoned to the electrode of the lambda sensor and the failure of the lambda sensor may be correctly determined.

Abstract: A removing poison apparatus of a lambda sensor includes: a lambda sensor detecting an oxygen concentration included in an exhaust gas; and a control unit differentiating a heating temperature and a heating time to remove poison of the lambda sensor depending on a lambda signal output from the lambda sensor in an overrun condition of the engine. By heating the lambda sensor depending on the magnitude of the lambda signal output from the lambda sensor, the lambda sensor may be inhibited from being degraded by combusting the material poisoned to the electrode of the lambda sensor and the failure of the lambda sensor may be correctly determined.

Abstract: An ignition coil assembly for an engine includes an ignition coil for supplying a voltage to a spark plug, a housing supporting the ignition coil, a driving circuit board disposed on the housing and including a circuit unit controlling the ignition coil, and a center cover protecting the driving circuit board. The ignition coil is electrically connected to the driving circuit board.

Abstract: An ignition coil assembly for an engine includes an ignition coil for supplying a voltage to a spark plug, a housing supporting the ignition coil, a driving circuit board disposed on the housing and including a circuit unit controlling the ignition coil, and a center cover protecting the driving circuit board. The ignition coil is electrically connected to the driving circuit board.