Abstract: A filter capable of automatically cutting off gas flow, including: a housing (1) with a cavity, a filter element (2) disposed in the cavity, a blocking mechanism disposed in the flow channel (13), and an unblocking member (23). The housing is provided with an inlet end (11) and an outlet end (12). The filter element (2) is provided with a first chamber (21). A second chamber (23) is formed between the filter element (2) and the housing (1), and a flow channel (13) is disposed between the first chamber (21) and the inlet end (11). The blocking mechanism has an unblocking state in which the blocking mechanism is cooperated with the unblocking member (23) to allow a gas flow to flow out of the flow channel (13), and a blocking state in which the blocking mechanism prevents the gas flow from flowing out of the flow channel (13).

Abstract: The present disclosure relates to the field of wheat genetic breeding technology, and provides a gene TaCaM-3A for seed dormancy in wheat and application thereof. The full-length sequence of gene TaCaM-3A is shown in SEQ NO.1. The present disclosure identifies a gene TaCaM-3A that negatively regulating seed dormancy and sprouting resistance. Functional loss of this gene can significantly enhance sprouting resistance of wheat varieties, and increase grain length, grain width, and yield per plant. This gene has the potential to simultaneously improve sprouting resistance and yield of wheat, and has significant breeding application value in wheat sprouting-resistance breeding.

Abstract: An object detection model and relationship prediction model are jointly trained with parameters that may be updated through a joint backbone. The offset detection model predicts object locations based on keypoint detection, such as a heatmap local peak, enabling disambiguation of objects. The relationship prediction model may predict a relationship between detected objects and be trained with a joint loss with the object detection model. The loss may include terms for object connectedness and model confidence, enabling training to focus first on highly-connected objects and later on lower-confidence items.

Abstract: The present disclosure relates to a composition comprising one or more ribonucleic acids (RNAs) encoding virus-like particle (VLP) forming elements of an influenza virus. The present disclosure further provides uses of the composition.

Abstract: A semiconductor die is separated from a semiconductor wafer using a method that involves performing a partial cut on the semiconductor wafer, applying tape lamination to a front side of the semiconductor wafer, grinding a back side of the semiconductor wafer, mounting the semiconductor wafer to a die attach film (DAF) layer, removing the tape lamination from the front side of the semiconductor wafer, and performing a DAF-die separation operation to separate the semiconductor die from the adjacent semiconductor die. A DAF laser is not used during the method of separating a semiconductor die from a semiconductor wafer. The front side is the side of the semiconductor wafer where integrated circuits are exposed. The partial cut is between the semiconductor die and an adjacent semiconductor die. The back side is opposite of the front side and the back side is a silicon layer of the semiconductor die.

Abstract: An example operation may include at least one of receiving, from a source dataset, a plurality of labeled images, receiving, from a target dataset, a plurality of images associated with a different visual domain, extracting, from each of the plurality of labeled images and each of the plurality of images from the target dataset, one or more feature representations indicative of visual characteristics, grouping the plurality of labeled images into a plurality of image clusters based on similarity among the one or more feature representations, comparing the one or more feature representations of each of the plurality of image clusters to the one or more feature representations of the plurality of images from the target dataset to determine a similarity ranking for each image cluster, selecting, from the plurality of image clusters, a subset of labeled images based on the similarity ranking and a selection limit, and storing the subset of labeled images in a memory.

Abstract: An example operation may include one or more of retrieving an annotated source dataset from a storage via a software application, retrieving a non-annotated target dataset from the storage via the software application, identifying a subset of data from the annotated source dataset, wherein the subset is configured to include source dataset data that is similar to the non-annotated target dataset, reducing the subset of data from the annotated source dataset by using a classifier to remove redundant data from the subset of data from the annotated source dataset, and classifying data from the non-annotated target dataset by a trained artificial intelligence (AI) model.

Abstract: An example operation may include at least one of determining, by a transformer encoder trained on annotated image-text data, first latents for a first dataset stored in a memory, and second latents for a second dataset stored in the memory, generating a similarity matrix based on comparisons between the first latents and the second latents, constructing a graph comprising nodes corresponding to the first latents and edges based on pairwise similarity exceeding a threshold, identifying connected components in the graph and selecting, from each component, at least one latent having a highest score from a classifier trained to approximate divergence between the first dataset and the second dataset, forming a reduced dataset comprising the at least one latent, providing the reduced dataset to a model training module, and training an image classifier using the reduced dataset and the second dataset.

Abstract: An example operation may include at least one of converting an annotated dataset loaded from a storage into a first set of latents, converting a non-annotated dataset loaded from the storage into a second set of latents creating an aligned subset of data from the annotated dataset comprising: clustering the first set of latents into a plurality of clusters, determining a discrepancy score for each cluster in the plurality of clusters and the second set of latents, creating a refined subset of data from the annotated dataset by including at least one data from each cluster of the plurality of clusters, wherein adding the at least one data lowers the discrepancy score of the refined subset of data and the second set of latents, determining a similarity score between latents in the first set of latents and the second set of latents, wherein the aligned subset of data is created from the annotated dataset by parsing the refined subset into pairs of latents and for each of the pairs of latents, including a latent

Abstract: A coalescing filter element, including an inner coalescing component configured to captured a large amount of liquid in gas, and an outer coalescing component configured to coalesce and filter a small amount of liquid remaining in the gas. The inner coalescing component and the outer coalescing component are cylindrical structures disposed in a vertical direction and opened at two ends. The outer coalescing component is sleeved on an outer side of the inner coalescing component, and an annular drainage space is formed between the inner coalescing component and the outer coalescing component. A top end cap is provided on top ends of the inner coalescing component and the outer coalescing component. A bottom end cap is provided on bottom ends of the inner coalescing component and the outer coalescing component. The bottom end cap is provided with a gas inlet communicated with an interior of the inner coalescing component.

Abstract: An object detection model and relationship prediction model are jointly trained with parameters that may be updated through a joint backbone. The offset detection model predicts object locations based on keypoint detection, such as a heatmap local peak, enabling disambiguation of objects. The relationship prediction model may predict a relationship between detected objects and be trained with a joint loss with the object detection model. The loss may include terms for object connectedness and model confidence, enabling training to focus first on highly-connected objects and later on lower-confidence items.

Abstract: The present application proposes a content addressable memory including: memory cells coupled with each word line and a bit line bus; a timing control circuit configured to determine an operation mode including read, write, and compare modes and generate a control signal for a corresponding operation mode according to a received bus command; an address line control circuit configured to activate a corresponding word line and the bit line bus based on the control signal; and a processing and output circuit configured to read out data stored in the memory cells from the bit line bus in the read mode, write data to be written from the bit line bus to the memory cells in the write mode, and compare data to be compared inputted from the bit line bus with data already stored in the memory cells and output a compare result in the compare mode.

Abstract: Implementations of the present disclosure provide a memory structure, a memory, and a memory system. The memory structure includes: a plurality of memory transistors arranged in an array along a first direction and a second direction; and a plurality of bit line groups extending along the first direction and arranged along the second direction, each of which includes a first, second, and third bit lines extending along the first direction and arranged along a third direction; wherein the memory transistor is connected to any one of the first, second, and third bit lines in the same bit line group, or the memory transistor is not connected to the first, second, and third bit lines in the same bit line group; the first direction intersects with the second direction, and both of the first and second directions are perpendicular to the third direction.

Abstract: A switchable two-stage coalescence separation system, including a coalescer housing, a plurality of two-stage filter elements, and a particle detector. A lower portion and an upper portion of each of the two-stage filter elements are located in a lower chamber and an upper chamber of the coalescer housing, respectively. Two gas inlet branch pipes are communicated with the lower chamber and the upper chamber, respectively and are connected to a gas inlet main pipe through a first multi-way valve. The particle detector is disposed on the gas inlet main pipe. Two outlet branch pipes are communicated with the lower chamber and the upper chamber, respectively and are connected to a gas outlet main pipe through a second multi-way valve.

Abstract: The present disclosure provides a spherical medical tantalum alloy powder, a preparation method therefor and a use thereof, the preparation method comprising the following steps: S1. conducting preliminary plasma spheroidization on a pure tantalum powder to obtain a spherical pure tantalum powder; S2. mixing the spherical pure tantalum powder obtained from S1 with a pure copper particle and a pure silver particle, and then subjecting the mixture to a planetary ball milling to obtain a mixed metal powder; and S3. conducting plasma spheroidization on the mixed metal powder obtained in S2, and then sieving and filtering to obtain the spherical medical tantalum alloy powder. In the present disclosure, through the preparation method combining two-stage plasma spheroidization technology with planetary ball milling technology, the burnout of copper and silver can be reduced, and the spheroidization rate of the tantalum alloy powder can be further improved.

Abstract: Provided is a starting material composition suitable for producing rigid isocyanurate foam excellent in both energy absorption performance and sound absorption performance, while exhibiting excellent foam moldability, appearance, and air permeability. The starting material composition for rigid isocyanurate foam according to the present disclosure contains a polyol composition and a polyisocyanate. The polyol composition contains a trifunctional or higher-functional polyether polyol (C) having a hydroxy value of 300 mg KOH/g or more, a polyether polyol (A) different from polyether polyol (C) having an ethylene oxide content, based on the total amount of alkylene oxide of polyether polyol (A), of 50 mass % or more, and a polyether polyol (B) different from polyether polyol (C) having an ethylene oxide content, based on the total amount of alkylene oxide of polyether polyol (B), of 10 mass % or less.

Abstract: An image retrieval system receives an image for which to identify relevant images from an image repository. Relevant images may be of the same environment or object and features and other characteristics. Images in the repository are represented in an image retrieval graph by a set of image nodes connected by edges to other related image nodes with edge weights representing the similarity of the nodes to each other. Based on the received image, the image traversal system identifies an image in the image retrieval graph and alternatively explores and traverses (also termed “exploits”) the image nodes with the edge weights. In the exploration step, image nodes in an exploration set are evaluated to identify connected nodes that are added to a traversal set of image nodes. In the traversal step, the relevant nodes in the traversal set are added to the exploration set and a query result set.

Abstract: A method and a device for driving a hybrid electric vehicle, and a vehicle is provided. The method includes: driving, if the vehicle demand torque is greater than a minimum value of the engine economic zone and less than a maximum value of the power assistance reserve zone, the hybrid electric vehicle in a parallel drive mode; or maintaining, if the vehicle demand torque continues to increase to the maximum value of the power assistance reserve zone, the hybrid electric vehicle being driven in the parallel drive mode when the hybrid electric vehicle is in the parallel drive mode and a torque provided by the parallel drive mode under a current drive mode is greater than a maximum torque provided by the drive motor.

Abstract: Disclosed are a high-entropy alloy (HEA) coating and a preparation method and use thereof. Laser cladding is conducted with an HEA powder to obtain the HEA coating, where the HEA is a FeCoCrNiAl0.5Ti0.5 alloy, and the HEA includes the following chemical components in atomic percentage: Al: 10.01% to 12.30%, Co: 18.1% to 22.5%, Cr: 18.05% to 20.12%, Fe: 18.77% to 21.02%, Ni: 19.21% to 20.99%, and Ti: 8.43% to 11.5%. The HEA material with high hardness and wear resistance provided by the present disclosure is suitable for laser cladding of a surface of a precision mold, an offshore component, or a drilling rod. A powder is prepared from the above alloy components and then prepared into a corresponding HEA coating with high strength, high hardness, and prominent wear resistance through laser cladding. The material has prominent weldability and is a special nickel-based HEA material suitable for laser additive manufacturing.

Abstract: A computing system may access first alpha values associated with first pixels in a first pixel region of an image and determine a bit budget for encoding the first alpha values. The computing system may then select a first alpha-encoding mode for the first alpha values to reflect a determination that the first alpha values are all fully transparent or all fully opaque, and encode the first alpha values by storing the selected first alpha-encoding mode as part of a metadata without using the bit budget to encode the first alpha values individually. The computing system may then update a record of unallocated bits available for allocation based on the bit budget unused in the encoding of the first alpha values, and allocate, based on the record of unallocated bits, bits to encode a set of alpha values different from the first alpha values.