Abstract: Systems and methods for entity classification via an artificial intelligence model using a defined taxonomy framework are described. Entity classification includes generating a first query for a classification model, the first query including a first set of options for classification of an entity at a first classification granularity level of a taxonomy framework, providing the first query comprising the first set of options for classification to the classification model, receiving, from the classification model, a selection of one or more options of the first set of options for classification, and determining a classification of the entity based at least in part on the selection of the one or more options of the first set of options.

Abstract: An implantable device or implant is configured to be positioned within a native heart valve to allow the native heart valve to form a more effective seal. In some examples, a plurality of sutures is employed to secure frame portions of the implantable device to a connector of the implantable device. Each of the sutures can be passed through one or more eyelets of the frame portions and the connector. In some examples, an implantable device includes an actuation cap having a flexible portion or a biasing element positioned between an end portion of the actuation cap and an anchor portion of the implantable device. The flexible portion or the biasing element flex or compress responsive to forces between the connector and the actuation cap.

Abstract: Embodiments of this application provide a method and an apparatus for resolving an intent conflict that includes: receiving conflict resolution policy information indicating a threshold and a first operation, the first operation includes: when a first preset condition is met, using an alternative solution for achieving a first intent, where the first preset condition includes that a first impact value of a value that is of a first KPI and that corresponds to the solution for a value that is of the first KPI and that corresponds to a second intent target is less than or equal to the threshold; and performing conflict resolution on the solution based on the conflict resolution policy information.

Abstract: An intent processing method. A first device sends first information to a second device, wherein the first information indicates impact on at least one KPI when a solution of a first intent is executed. After receiving the first information, the second device may determine the execution suggestion for the first intent based on the first information. The second device sends the information that indicates an execution suggestion for the first intent to the first device.

Abstract: Systems and methods for entity classification via an artificial intelligence model using a defined taxonomy framework are described. Entity classification includes identifying information associated with a entity, generating, by a processing device, a first query comprising the information associated with the entity and a first set of options for classification of the entity at a first level of classification granularity of a taxonomy framework, and providing the first query to a classification model. The method further includes receiving, from the classification model, a selection of a first option of the first set of options for classification and determining a classification of the entity based at least in part on the selection of the first option of the first set of options.

Abstract: Embodiments of this application provide an intent pre-evaluation method and apparatus. The method includes: A first network element receives an intent target and a pre-evaluation type from a second network element; receives first information from a third network element, where the first information includes one or more of the following: operation impact information or first KPI impact information, the operation impact information indicates mutual exclusion information of the first operation, and the first KPI impact information indicates impact of the first operation on a KPI associated with the intent target or impact of the first operation on the network KPI; and determines, based on the first information, result information corresponding to the pre-evaluation type, and sends the result information to the second network element, where the result information includes pre-evaluation information of an intent achievement result or pre-evaluation information of impact on a KPI.

Abstract: A communication method and apparatus for network energy saving. A second device is a requester device of energy saving analysis, and a first device is an execution device of energy saving analysis. The first device receives, from the second device, first information including an energy saving target of a network, and sends an energy saving analysis result corresponding to the energy saving target to the second device. The energy saving target includes at least one of energy consumption reduction and energy efficiency improvement. The energy saving analysis result corresponds to the energy saving target of the network.

Abstract: This application provides a method and a related communication apparatus applicable to the communication field. In the method, a network device receives a first notification message sent by a terminal device, where the first notification message indicates that the terminal device has completed training on a target model, and the first notification message includes a model name corresponding to the target model. The network device then determines policy information based on the uplink channel state information of the terminal device; and sends a lightweight indication message to the terminal device.

Abstract: An implantable device is configured to be positioned within a native heart valve to allow the native heart valve to form a more effective seal. The implantable device includes a cover. The cover can be attached to one or more sleeves, can have a portion that has a lower coefficient of friction, and/or can have separate pieces that cover different portions of the implantable device.

Abstract: This application provides a model data sending method and an apparatus. The method can reduce an accuracy loss of model data, and includes: A first device determines second information based on first information, where the second information is used by a second device to quantize first model data, the first information includes an evaluation loss corresponding to a current round of training, the second information includes a quantization error threshold, and the first model data is model data that is after the current round of training; the first device sends the second information to the second device; and the first device receives a first message sent by the second device, where the first message includes quantized first model data and first quantization configuration information.

Abstract: A model distillation training method provides for establishment of a distillation training communication connection with a second device prior to performing distillation training on a neural network model. Based on exchange of distillation training information between the first device and the second device, the second device configures a first reference neural network model by using first configuration information sent by the first device. After configuring the first reference neural network model, the second device performs operation processing on first sample data in first data information based on the configured first reference neural network model by using the first data information to obtain first indication information, and sends the first indication information to the first device. The first device trains, by using the first indication information, a first neural network model designed by the first device.

Abstract: The present invention relates to an antigen-binding protein, or an antigen-binding fragment thereof, comprising (i) a heavy chain variable domain comprising a VHCDR1 having the amino acid sequence GYSITSGYSWH; a VHCDR2 having the amino acid sequence YIHYSGSTKYNPSLKS and a VHCDR3 having the amino acid sequence GSNYGFDY; and (ii) a light chain variable domain comprising a VLCDR1 having the amino acid sequence KSSQSLLYSNDQKNYLA, a VLCDR2 having the amino acid sequence WASIRES, and a VLCDR3 having the amino acid sequence QQYYIYPLT.

Abstract: A spiral-orifice ceramic filter for metal casting, including spiral channels and two drain openings, where the spiral channels are distributed in a ceramic substrate in a staggered manner. By adoption of the spiral channel structure, molten metal may rotate to generate a centrifugal force while flowing forwards so as to promote separation of inclusions. The spiral-orifice ceramic filter for metal casting includes the following components: 90-95 wt % of MgO, 4-8 wt % of SiO2 and 2-4 wt % of ZrO2. Therefore, the spiral-orifice ceramic filter for metal casting has high strength under normal temperature and optional thermal impact resistance under high temperature, and may tolerate the impact of molten metal at 1700° C. or higher without break. The ceramic substrate and the spiral channel are superficially coated with one layer of functional oxide prepared from CaO.2Al2O3, CaO.6Al2O3, Al2O3, TiO2, or Re2O3.

Abstract: A mussel soybean paste and a preparation method thereof are provided, the preparation method includes: sterilizing soaked soybeans, adding mussel cooking liquor hydrolysate, and then inoculating Aspergillus oryzae for making koji to obtain soybean koji; sterilizing soaked rice, and inoculating Aspergillus oryzae for making koji to obtain rice koji; mixing the rice koji with the soybean koji to obtain a koji material; adding the koji material and salt into mussel meat for anaerobic fermentation, grinding an obtained mussel soybean paste semi-finished product by using a colloid mill, and conducting sterilizing to obtain the mussel soybean paste. The mussel soybean paste which is natural, nutritional and good in flavor can be prepared, and the added value of mussels is increased.

Abstract: An image detection device and an image detection method are provided. The image detection method includes: obtaining an image, where the image includes an object; adjusting a first size of the image to generate an adjusted image; generating a first divided image and a second divided image according to the image; and detecting the object in the image based on a plurality of input images, where the plurality of input images includes the first divided image, the second divided image, and the adjusted image.

Abstract: An object detection device and an object detection method based on a neural network are provided. The object detection method includes: receiving an input image and identifying an object in the input image according to an improved YOLO-V2 neural network. The improved YOLO-V2 neural network includes a residual block, a third convolution layer, and a fourth convolution layer. A first input of the residual block is connected to a first convolution layer of the improved YOLO-V2 neural network, and an output of the residual block is connected to a second convolution layer of the improved YOLO-V2 neural network. Here, the residual block is configured to transmit, to the second convolution layer, a summation result corresponding to the first convolution layer. The third convolution layer and the fourth convolution layer are generated by decomposing a convolution layer of an original YOLO-V2 neural network.

Abstract: An image detection device and an image detection method are provided. The image detection method includes: obtaining an image, where the image includes an object; adjusting a first size of the image to generate an adjusted image; generating a first divided image and a second divided image according to the image; and detecting the object in the image based on a plurality of input images, where the plurality of input images includes the first divided image, the second divided image, and the adjusted image.

Abstract: Embodiments of the present invention are directed to a computer-implemented method for translating a content item. Aspects include obtaining a user profile for a reader of the content item, the user profile comprising a native language of the reader, a second language of the reader and skill level associated with the second language. Aspects also include obtaining the content item, wherein the content item is in the second language. Aspects further include creating a translated content item by determining, for each phrase in the content item, a confidence score that the reader will understand the phrase based at least in part on the user profile and replacing each phrase in the content item having the confidence score less than a threshold level with an alternate phrase. Aspects also include presenting the translated content item to the reader.

Abstract: An object detection device and an object detection method based on a neural network are provided. The object detection method includes: receiving an input image and identifying an object in the input image according to an improved YOLO-V2 neural network. The improved YOLO-V2 neural network includes a residual block, a third convolution layer, and a fourth convolution layer. A first input of the residual block is connected to a first convolution layer of the improved YOLO-V2 neural network, and an output of the residual block is connected to a second convolution layer of the improved YOLO-V2 neural network. Here, the residual block is configured to transmit, to the second convolution layer, a summation result corresponding to the first convolution layer. The third convolution layer and the fourth convolution layer are generated by decomposing a convolution layer of an original YOLO-V2 neural network.