Abstract: A method for problem inference based on multi-modal generative artificial intelligence includes receiving question information including an image and text, generating formal languages by parsing the image and text of the question information, respectively, based on a pre-constructed problem solving template, generating text-based intermediate inference information for the question information by inputting the generated formal language to a formal language inference unit, generating image-based inference information by inputting the text-based intermediate inference information, the text included in the question information (hereinafter referred to as “text question information”), and the image included in the question information (hereinafter referred to as “image question information”) to a multi-modal image generation model, and generating text-based inference information by inputting the text-based intermediate inference information, the image-based inference information, and the text question information

Abstract: An air cleaner includes a case, an air inlet, an air outlet, a grill covering the air inlet to filter airborne oil droplets, a fan to suction air through the air inlet and discharge air out of the air outlet, a roll filter provided between the grill and the fan to filter oil mist, and a filter roller to cut the roll filter. The roll filter includes a roller and a filter sheet wound around the roller, and the filter sheet may be unwound to extend behind the grill. When the filter sheet becomes dirty, the roll filter is further unwound so that a clean section is provided behind the grill, and the filter cutter may cut off the dirty section. During cutting, a clamp holds down the filter sheet. The case includes a slit through which the dirty section of the filter sheet passes for discarding.

Abstract: The present disclosure relates to a thin film forming method, and more particularly, to a thin film forming method for forming a gallium nitride thin film. In accordance with an exemplary embodiment, a thin film forming method includes: loading a substrate into a process space of a chamber; and forming a gallium nitride thin film on the substrate, and the forming of the gallium nitride thin film includes: supplying a source gas containing gallium onto the substrate; supplying a reactant gas containing nitrogen onto the substrate; and activating and supplying a post-treatment gas containing hydrogen onto the substrate.

Abstract: Provided is a method for cleaning a substrate processing apparatus. The method includes loading a substrate into a chamber, injecting a gas containing at least one of Zn, Ga, In, or Sn into the chamber to deposit a thin film on the substrate, unloading the substrate to the outside of the chamber, injecting a cleaning gas containing Br into the chamber, and exhausting byproducts generated through a reaction between impurities deposited inside the chamber in addition to the substrate and the cleaning gas in the depositing of the thin film. Therefore, the inside of the apparatus may be cleaned at relatively low temperature. That is, impurities having the form of the thin film, which are deposited on a surface of a component or structure installed inside the apparatus may be delaminated from the surface.

Abstract: A semiconductor wafer sample pre-processing system improves productivity and process reliability by automating all process from cutting a into a plurality of wafer samples through polishing of the wafer samples.

Abstract: Disclosed are a raw material supply method including vaporizing a raw material in a canister, discharging the vaporized raw material, measuring an inner temperature of the canister, calculating a calculated temperature by using the inner temperature, and compensating a variation of the inner temperature by heating a heating unit disposed on the canister at the calculated temperature and a raw material supply apparatus applied to the method for supplying the raw material. The raw material supply method and apparatus may stably supply the vaporized raw material to a process space.

Abstract: Provided is a method of generating a language model using crossmodal information. The method includes: receiving language-based first modality information and non-language-based second modality information; converting the first modality information into a first byte sequence; converting the second modality information into a second byte sequence; converting the first and second byte sequences into a first embedding vector and a second embedding vector by applying an embedding technique for each modality; generating semantic association information between first and second modality information by inputting the first and second embedding vectors to a crossmodal transformer; and learning the language model by setting the generated semantic association information as training data.

Abstract: The present invention relates to an apparatus and method for automatically generating machine reading comprehension training data, and more particularly, to an apparatus and method for automatically generating and managing machine reading comprehension training data based on text semantic analysis. The apparatus for automatically generating machine reading comprehension training data according to the present invention includes a domain selection text collection unit configured to collect pieces of text data according to domains and subjects, a paragraph selection unit configured to select a paragraph using the pieces of collected text data and determine whether questions and correct answers are generatable, and a question and correct answer generation unit configured to generate questions and correct answers from the selected paragraph.

Abstract: Disclosed herein are an apparatus and method for providing a search service based on important sentences. The apparatus for providing a search service based on important sentences includes memory in which at least one program and a previously trained word importance measurement model are recorded and a processor for executing the program. The program may include a word importance measurement unit for measuring the importance of each of multiple words included in input text in the corresponding input text based on the word importance measurement model and a sentence importance measurement unit for measuring the importance of each of at least one sentence included in the text based on the measured importance of each of the multiple words.

Abstract: An air cleaner includes a case, an air inlet, an air outlet, a grill covering the air inlet to filter airborne oil droplets, a fan to suction air through the air inlet and discharge air out of the air outlet, a roll filter provided between the grill and the fan to filter oil mist, and a filter roller to cut the roll filter. The roll filter includes a roller and a filter sheet wound around the roller, and the filter sheet may be unwound to extend behind the grill. When the filter sheet becomes dirty, the roll filter is further unwound so that a clean section is provided behind the grill, and the filter cutter may cut off the dirty section. During cutting, a clamp holds down the filter sheet. The case includes a slit through which the dirty section of the filter sheet passes for discarding.

Abstract: An air cleaner includes a case, an air inlet, an air outlet, a grill covering the air inlet to filter airborne oil droplets, a fan to suction air through the air inlet and discharge air out of the air outlet, a roll filter provided between the grill and the fan to filter oil mist, and a filter roller to cut the roll filter. The roll filter includes a roller and a filter sheet wound around the roller, and the filter sheet may be unwound to extend behind the grill. When the filter sheet becomes dirty, the roll filter is further unwound so that a clean section is provided behind the grill, and the filter cutter may cut off the dirty section. During cutting, a clamp holds down the filter sheet. The case includes a slit through which the dirty section of the filter sheet passes for discarding.

Abstract: An object of the present invention is to provide a heating carrier that does not heat all of exhaust gas flowing into a catalyst converter, but directly supplies, to a catalyst layer, thermal energy in the form of an instantaneous pulse to effectively activate a catalyst during a cold start-up period, and thus may reduce emission pollutants with a small amount of energy, and an exhaust gas reduction carrier having the heating carrier. In order to accomplish the object, the heating carrier of the present invention may include a main body of which the inside is formed to have a honeycomb structure, the main body being formed of a conductive ceramic material that is a nonmetallic heating element; and a catalyst layer formed by coating a first catalyst on a surface of the main body.

Abstract: An air cleaner includes a case, an air inlet, an air outlet, a grill covering the air inlet to filter airborne oil droplets, a fan to suction air through the air inlet and discharge air out of the air outlet, a roll filter provided between the grill and the fan to filter oil mist, and a filter roller to cut the roll filter. The roll filter includes a roller and a filter sheet wound around the roller, and the filter sheet may be unwound to extend behind the grill. When the filter sheet becomes dirty, the roll filter is further unwound so that a clean section is provided behind the grill, and the filter cutter may cut off the dirty section. During cutting, a clamp holds down the filter sheet. The case includes a slit through which the dirty section of the filter sheet passes for discarding.

Abstract: An air cleaner includes a case, an air inlet, an air outlet, a grill covering the air inlet to filter airborne oil droplets, a fan to suction air through the air inlet and discharge air out of the air outlet, a roll filter provided between the grill and the fan to filter oil mist, and a filter roller to cut the roll filter. The roll filter includes a roller and a filter sheet wound around the roller, and the filter sheet may be unwound to extend behind the grill. When the filter sheet becomes dirty, the roll filter is further unwound so that a clean section is provided behind the grill, and the filter cutter may cut off the dirty section. During cutting, a clamp holds down the filter sheet. The case includes a slit through which the dirty section of the filter sheet passes for discarding.

Abstract: Disclosed herein are an apparatus and method for providing a search service based on important sentences. The apparatus for providing a search service based on important sentences includes memory in which at least one program and a previously trained word importance measurement model are recorded and a processor for executing the program. The program may include a word importance measurement unit for measuring the importance of each of multiple words included in input text in the corresponding input text based on the word importance measurement model and a sentence importance measurement unit for measuring the importance of each of at least one sentence included in the text based on the measured importance of each of the multiple words.

Abstract: An electrochemical cell is prepared from a silicon-based anode active material and a polyimide-based binder prepared by curing polyamic acid with heat and/or with a catalyst. The silicon-based material may be silicon suboxide. Anodes prepared with combinations of an elastic polyimide-based binder and anode active material improve specific capacity, cycle characteristics, and electrical properties in secondary lithium batteries.

Abstract: Provided are porous inorganic particles which may be added to an inner portion and/or an outer portion of a secondary battery including a nickel-rich lithium composite transition metal oxide active material in which nickel accounts for 50 mol % or more of all metals excluding lithium, wherein the porous inorganic particles include oxides of four or more types of metals selected from the group consisting of Al, Si, Na, Ca, and K. When these porous inorganic particles are used, swelling of a secondary battery may be suppressed, and a nickel-rich lithium composite transition metal oxide may be stably used as an active material of a secondary battery.

Abstract: Provided are a system and a method for providing service to recommend customized exercise used for autonomous vehicle. The system and method are configured to determine whether the user does the exercise safely and freely during autonomous control of the vehicle through the monitoring system in the autonomous vehicle and recommend the customized exercise used for the autonomous vehicle according to a situation of the user and the situation of the autonomous control and to determine whether the user does the exercise safely and freely during the autonomous control of the vehicle through the monitoring system in the autonomous vehicle and recommend the autonomous vehicle in which the user can do the recommended exercise according to the situation of the user and the situation of the autonomous control.

Abstract: A method of selecting a call connection device according to perception of surrounding conditions is disclosed. The method includes sensing an external electronic device paired with a user's portable device while waiting to receive a call, collecting first information on the paired external electronic device, recognizing at least one of first peripheral information recognizing surrounding conditions of the external electronic device and second peripheral information recognizing surrounding conditions of the portable device based on the collected first information, and learning the recognized first peripheral information or the recognized second peripheral information and recommending one of the portable device and the external electronic device suitable for surrounding conditions of the user based on a result of the learning to connect a call with the other party. Hence, the user can prevent the exposure of personal privacy by recommending the call connection device suitable for user's surroundings.

Abstract: The present invention relates to an apparatus and method for automatically generating machine reading comprehension training data, and more particularly, to an apparatus and method for automatically generating and managing machine reading comprehension training data based on text semantic analysis. The apparatus for automatically generating machine reading comprehension training data according to the present invention includes a domain selection text collection unit configured to collect pieces of text data according to domains and subjects, a paragraph selection unit configured to select a paragraph using the pieces of collected text data and determine whether questions and correct answers are generatable, and a question and correct answer generation unit configured to generate questions and correct answers from the selected paragraph.