Abstract: A method for generating an image. The method includes: providing a randomly drawn image or representation thereof as input of a sequence of layers of a neural network which includes a cross-attention layer; providing a first input, which is a representation of the input of the sequence determined by layers preceding the cross-attention layer or is the input to the sequence of layers, and providing a text embedding characterizing a description of the image to be generated as a second input; determining, by the cross-attention layer, an attention map based on the first and second inputs; optimizing the input provided to the sequence of layers based on a loss function which includes a term characterizing a negative total variation of the attention map; determining an output of the sequence of layers using the optimized input; and determining the image based on the determined output.

Abstract: A method for training a task neural network for a task to be performed on an input including images. The method includes: providing 2D training images labelled with ground truth; expanding the training images into 3D representations; processing, by the task neural network, the 2D training images into task outputs; processing, by an auxiliary neural network, the 3D representations into auxiliary outputs; rating, by a task loss function, a deviation of the task output for each training image from the ground truth with which it is labelled; rating, by an auxiliary loss function, a plausibility of an outcome of the task neural network produced from at least one training image with the an outcome of the auxiliary neural network produced from the corresponding 3D representation; aggregating values of the task loss function and the auxiliary loss function; and optimizing parameters that characterize the behavior of the task neural network.

Abstract: This application provides a shooting method and a related device. The shooting method may be applied to an electronic device with a camera lens. In the shooting method, the electronic device may output a RAW image through the camera lens, and divide the image into three paths of data streams through an image processor. The three paths of data streams are respectively data streams of a panorama path, a focus tracking path, and a detection path respectively. After stabilization processing, the data stream of the panorama path is used to display and preview a large window of a camera application interface. After human body detection and successful focus tracking, information about a human body cropping frame may be output, through the data stream of the detection path, to the focus tracking path for cropping a human body frame.

Abstract: Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive an indication of a configuration to perform cooperative radar sensing. The configuration may indicate resources for the first UE to use to communicate radar measurement reports. The UE may receive, from another UE on the identified resources, a radar measurement report that includes values for one or more radar measurement parameters for a first radar target and a time stamp associated with the values. The UE may identify a second set of values for the radar measurement parameters (e.g., its own measurements, or measurements received from a third UE in another radar measurement reports). The UE may generate a combined set of values for the radar measurement parameters for each time stamp by combining the first set of values and the second set of values.

Abstract: Methods, systems, and devices for wireless communications are described. A satellite may determine one or more sets set of resources for conveying reference signals to a user terminal. In one example, the satellite may determine a set of resources that includes multiple resource groups each associated with a respective frequency range. In another example, the satellite may determine multiples sets of resources and each set of resources may include resources within respective frequency range. Upon determining the set(s) of resources, the satellite may transmit an indication of the set(s) of resources to a user terminal.

Abstract: Disclosed embodiments relate to systems, methods, and computer readable storage media for performing dataset discovery. Some embodiments may include accessing a data repository having a plurality of tables having cell values arranged in one or more columns and one or more rows, generating serialized sequences of the cell values that correspond to particular columns of the plurality of tables, inputting the serialized sequences into a natural language model, converting, using the natural language model, the serialized sequences into contextualized embeddings associated with the plurality of tables, storing the contextualized embeddings associated with the plurality of tables in one or more vector indices, receiving a query table, or generating an output of one or more candidate tables from the plurality of tables that are unionable with the received query table.

Abstract: A method for wireless communication performed by a user equipment (UE) includes receiving a first message comprising a configuration for reconstructing a bandwidth part (BWP) for a network entity. The BWP may be used on a non-terrestrial beam of a non-terrestrial entity. The method also includes reconstructing a BWP of the BWPs based on the configuration. The method further includes switching from a current BWP to the reconstructed BWP to communicate with the network entity, e.g., non-terrestrial entity.

Abstract: A biological self-assembled nanocrystal injection having a lymphatic targeting function and an administration method for targeting a lymphatic system using the biological self-assembled nanocrystal injection. The injection comprises 0.05-5% of mitoxantrone and a salt thereof, 0.1-10% of an osmotic pressure regulator, 0.001-0.1% of a buffering agent, 0.01-0.1% of an antioxidant, 0.05-1% of an adsorbent, and 0-20% of a filler. The administration method is interstitial injection, and a lymph node will produce a blue color visible to the naked eye after passive targeting to the lymphatic system.

Abstract: Generating intermediate predictions in a trained machine learning model. The trained machine learning models may include one or more input layers, multiple intermediate layers, and one or more output layers, which generate an output. A computer-implemented method may perform classification using the trained machine learning model, which includes the steps of feeding input data to the trained machine learning model, propagating the input data through a part of the trained machine learning model, obtaining intermediate predictions from the layers in the part of the trained machine learning model, ensembling these intermediate predictions to obtain an ensemble prediction, and using the ensemble prediction as a substitute for the output of the trained machine learning model in the classification. The ensembling phase may include determining the ensemble prediction as a product of weighted versions of the obtained intermediate predictions, and having a normalized probability density.

Abstract: A device and computer implemented method for evaluating a digital image. The method includes providing the digital image, providing a first part of a predetermined model, wherein the predetermined model is configured for determining a semantic segmentation of the digital image with a second part of the predetermined model, wherein the first part is configured to determine a feature depending on the digital image, wherein the second part is configured to determine the semantic segmentation depending on the feature, wherein the method comprises determining the feature depending on the digital image with the first part, providing a set of quantizations for quantizing the feature, determining the quantization of the feature depending on the set of quantizations and depending on the feature, determining a quantization error depending on the feature and the quantization, and evaluating the digital image depending on the quantization error.

Abstract: Certain aspects of the present disclosure provide techniques for wireless communication. A method that may be performed by a base station (BS), the method including determining a plurality of root sets, wherein each root set of the plurality of root sets comprise at least a first root and a second root. In some examples, the method includes transmitting signaling that indicates the plurality of root sets.

Abstract: A filter assembly, a vacuum cleaner and a cleaning set are provided. The filter assembly comprises a cyclone separator and a high efficiency particulate air filter (HEPA) assembly; the cyclone separator comprises a filter support and a filter screen, an air guide structure which protrudes outwards in a radial direction and spirally extends towards one side of the filter screen is formed on a side wall of one end of the filter support, and a plurality of through first air outlet holes are formed in the air guide structure; and the HEPA assembly comprises an HEPA support and an HEPA body, a plurality of second air outlet holes are formed in the HEPA support, the HEPA body is arranged on the HEPA support and comprises a plurality of folds, and an airflow circulation channel in an axis direction of the cyclone separator is formed between any two adjacent folds.

Abstract: Disclosed are techniques for positioning. A receiver receives complementary timing information associated with the positioning reference signals (PRS) of a repetition of a PRS sequence. The complementary timing information distinguishes a repetition of a PRS sequence from remaining repetitions of the PRS sequence. Based on the complementary timing information associated with different PRS from different non-terrestrial vehicles indicating that the PRS were transmitted during the same radio frame, the receiver can determine an observed time difference of arrival (OTDOA) between the PRS received from the respective non-terrestrial vehicles.

Abstract: A backlight module includes a first substrate, a control circuit, a plurality of light source points, a filling layer, and a reflection structure. The control circuit is arranged on a first side of the first substrate. The plurality of light source points are arranged on a side of the control circuit away from the first substrate, include at least one type of LED, and are controlled by the control circuit to emit light to a side of the LED away from the first substrate. The filling layer covers the plurality of light source points and a gap between neighboring light source points. The reflection structure is arranged on a side of the filling layer away from the plurality of light source points and configured to reflect the light emitted by the plurality of light source points.

Abstract: Methods, systems, and devices for wireless communications are described. A satellite may determine one or more sets set of resources for conveying reference signals to a user terminal. In one example, the satellite may determine a set of resources that includes multiple resource groups each associated with a respective frequency range. In another example, the satellite may determine multiples sets of resources and each set of resources may include resources within respective frequency range. Upon determining the set(s) of resources, the satellite may transmit an indication of the set(s) of resources to a user terminal.

Abstract: Methods, systems, and devices for wireless communication are described. A communication device (e.g., a vehicle) in wireless communications system (e.g., a cellular-vehicle-to-everything (V2X) system) may support adaptive radar transmissions based on information received in a public safety message. The communication device may use information included in the public safety message to adapt radar transmissions to enable timely detection of vulnerable road users (VRUs). In some examples, based on a location and a velocity estimate provided in the public safety message, the communication device may adjust the radar transmissions to experience a trade-off between range and velocity estimation performance. Additionally or alternatively, based on positional accuracy estimates provided in the public safety message, the communication device may adjust the radar transmissions to improve beamforming.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining a hardware datapath for a hardware accelerator computer chip.

Abstract: Provided is an antibody capable of specifically binding to CD47or an antigen binding fragment thereof. Also provided is a recombinant oncolytic virus, which is operably inserted with or comprises a gene coding sequence of an anti-CD47antibody or a CD47ligand, wherein the anti-CD47antibody comprises an Fc mutant having A330L/I332E mutations (ALIE antibody), i.e., the anti-CD47antibody is ?CD47-Fc(ALIE). Also provided are a preparation method for the recombinant oncolytic virus, a use of the recombinant oncolytic virus in preparation of anti-tumor drugs, and a vaccinia virus Tiantan strain capable of efficiently expressing an ?CD47-Fc(ALIE) gene.

Abstract: Methods, systems, and devices for wireless communications are described. A wireless device may implement one or more listen before talk (LBT) techniques. The wireless device may identify a set of transmission parameters of a signal for transmission from the wireless device. The wireless device may generate a first transmission waveform of the signal on a set of radio resources based on the set of transmission parameters. The wireless device may perform a LBT based at least in part on the generation of the first transmission waveform. In some cases, performing the LBT includes adjusting a detected signal associated with the set of radio resources based at least in part on the first transmission waveform. The wireless device may transmit the signal based on a result of the LBT.

Abstract: Disclosed are a data transmission method and apparatus, a non-transitory computer readable storage medium, and a computing device. The data transmission method is executed by a computing device, and includes: receiving first packets transmitted by a first device over multiple data channels; parsing the first packets to obtain address information of the first device, and performing aggregation on the first packets according to headers of the first packets, to obtain second packets, the designated address information comprising a virtual port assigned to the first device and associated with the address information of the first device; replacing source address information of the second packets with designated address information, to obtain third packets; and transmitting the third packets to a second device, the second device being a device that the first device needs to access.