Abstract: Provided herein are compounds, and pharmaceutically acceptable salts thereof, useful as KRAS inhibitors, methods of making and using the same (singly or in combination with additional agents), and pharmaceutical compositions thereof.

Abstract: Examples relate to fluid collection devices, and related systems and methods. In some embodiments, a fluid collection device includes a fluid impermeable barrier and a fluid permeable body. The fluid impermeable barrier may include a first end region, a second end region, a front side, and a back side. The fluid impermeable barrier defines an opening on the front side, one or more apertures on the back side distal to the opening and positioned between the first end region and the second end region, and a chamber within the fluid collection device. At least one of the first end region or the second end region may be substantially planar. The fluid permeable body is positioned within the chamber and extends across at least a portion of the opening.

Abstract: There are provided systems and methods for sentence level dialogue summaries using unsupervised machine learning for keyword selection and scoring. A service provider, such as an electronic transaction processor for digital transactions, may provide computing services to users, which may be used to engage in interactions with other users and entities. When utilizing these services, memoranda may be generated that includes text from different online interactions. To provide summarization and searching of the memoranda, the service provider may implement an unsupervised machine learning framework that utilizes machine learning models to perform text preprocessing, keyword extraction, and keyword weighting when ranking and outputting relevant keywords. Once extracted and ranked, the keywords may be used to provide different insights to the memoranda. The keywords may be selected based on domains and tasks and the framework may be made pluggable and customizable for different systems and search operations.

Abstract: Embodiments described herein provide a method of generating a multi-modal task output to a text instruction relating to inputs of multiple different modalities (e.g., text, audio, video, 3D). The method comprises receiving, via a data interface, a first input of a first modality, a second input of a second modality and the text instruction relating to the first and the second inputs; encoding, by a first multimodal encoder adapted for the first modality, the first input of the first modality into a first encoded representation conditioned on the text instruction; encoding, by a second multimodal encoder adapted for the second modality, the second input of the second modality into a second encoded representation conditioned on the text instruction; and generating, by a neural network based language model, the multi-modal task output based on an input combining the first encoded representation, the second encoded representation, and the text instruction.

Abstract: This disclosure is directed to methods and systems that establish a secure data channel between a host and a disaggregated hardware device (“DHD”) of a data center. The system comprises an initiator host that runs objects, such as virtual machines and containers. The host includes an initiator smart network interface card (“SNIC”). The initiator SNIC includes a virtual device, a trust platform module (“TPM”) and a security engine. The system also comprises a target host equipped with a DHD and a target SNIC. The target SNIC includes a TPM and a security engine. The TPM and the security engine of the initiator SNIC and the TPM and the security engine of the target SNIC establish a secure data channel between an object running on the host and the DHD.

Abstract: This application discloses a container management method, apparatus, and device. The method includes: obtaining a provisioning request that carries an identifier of a first user and an identifier of a first container image; querying and obtaining the identifier of the first container image in a container image information database of the first user; obtaining the container image based on the identifier of the first container image; and provisioning a container based on the container image.

Abstract: The present disclosure presents a device and method for monitoring and early warning of information perception regarding urban yellow mud water, featuring a rainfall collecting component, a video shooting component, a sediment content monitoring component, a water level monitoring component and a main controller. The video shooting component evaluates the long-term trend of yellow mud water production in a monitoring area by analyzing the color depth of captured water flow, sediment content in the pipeline's flow section, and meteorological rainfall data from the rainfall collecting component. Utilizing data on color, sediment content, flow section, and rainfall, this system achieves unsupervised real-time monitoring and precise early warning for yellow mud water. It provides a scientific basic for traceability, dynamic monitoring, and address challenges such as low measurement accuracy and omitted applicability found in existing methods.

Abstract: Certain aspects of the present disclosure provide techniques for beam selection for random access. A method that may be performed by a user equipment (UE) includes receiving, from a network entity, via a first beam, system information associated with at least a first plurality of beams that are within a coverage area of the first beam. The method also includes selecting a second beam among at least the first plurality of beams based at least in part on the system information and performing a random access procedure via the selected second beam.

Abstract: A method may include obtaining an original image. The method may also include determining a plurality of decomposition coefficients of the original image by decomposing the original image. The method may also include determining at least one enhancement coefficient by performing enhancement to at least one of the plurality of decomposition coefficients using a coefficient enhancement model. The method may also include generating an enhanced image corresponding to the original image based on the at least one enhancement coefficient.

Abstract: A method of training a neural network based three-dimensional (3D) encoder is provided. A first plurality of samples of a training dataset are generated using a first 3D model. An image generator with multi-view rendering is used to generate a plurality of two-dimensional (2D) images having different viewpoints of the first 3D model. A first language model is used to generate a plurality of texts corresponding to the plurality of 2D images respectively. A first text for a first image is generated by using one or more text descriptions generated by the first language model. A point cloud is generated by randomly sampling points in the 3D model. The first plurality of samples are generated using the plurality of 2D images, the corresponding plurality of texts, and the point cloud. The neural network based 3D encoder is trained using the training dataset including the first plurality of samples.

Abstract: A method and a system for medical imaging may be provided. A first optical image of a target subject that includes a target region to be scanned or treated by a medical device may be obtained. At least one body part boundary and at least one feature point of the target subject may be identified using at least one target recognition model from the first optical image. An image region corresponding to the target region of the target subject may be identified from the first optical image based on the at least one body part boundary and the at least one feature point. At least one first edge of the image region may be determined based on the at least one body part boundary, and at least one second edge of the image region may be determined based on the at least one feature point.

Abstract: Certain aspects of the present disclosure provide a method for wireless communications by a UE that generally includes transmitting a first sounding reference signal (SRS), from a first antenna port in a first symbol, on one or more tones determined by a first comb size and a first tone offset, transmitting a second SRS, from a second antenna port in the first symbol, on one or more tones determined by the first comb size and a second tone offset different than the first tone offset, transmitting a third SRS, from the first antenna port in a second symbol, on one or more tones determined by the first comb size and the second tone offset, and transmitting a fourth SRS, from the second antenna port in the second symbol, on one or more tones determined by the first comb size and the first tone offset.

Abstract: A thrust reversal emergency jettison device for an underwater vehicle includes a thrust reversal mechanism. The thrust reversal mechanism includes a casing and a gas reaction compartment provided in the casing; the gas reaction compartment communicates with a gas passage in a shell of the casing through a gas pipe; and a bottom end cover is provided at a bottom of the gas reaction compartment, and the bottom end cover is hermetically connected with the gas reaction compartment through a cylinder mechanism, and falls off automatically when a gas pressure in the gas reaction compartment increases. Compared with the prior art, the present invention can provide a reverse thrust for an underwater vehicle in an emergency, help the underwater vehicle to ascend, and improve safety.

Abstract: Aspects of the present disclosure provide techniques for uplink (UL) data channel design. An example method is provided for operations which may be performed by a first apparatus. The example method generally comprises determining a number of pilot symbols to transmit for one or more slots of a first subframe based, at least in part, on a coverage enhancement (CE) level, and transmitting at least one uplink data channel having the determined number of pilot symbols in the one or more slots of the first subframe.

Abstract: Wireless communications systems and methods related to communicating ACK/NACK feedbacks for multi-user and unicast physical down-link shared channel (PDSCH) communications are provided. A UE may receive a multi-user PDSCH communication including a first transport block (TB) for the UE and may receive a unicast PDSCH communication including a second TB for the UE. The UE may determine a hybrid automatic repeat request (HARQ) codebook for the first TB associated with the multi-user PDSCH communication and the second TB associated with the unicast PDSCH communication. The UE may transmit a first acknowledgement/negative-acknowledgment (ACK/NACK) feedback for the first TB and a second ACK/NACK feedback for the second TB based on the determined HARQ codebook.

Abstract: A user equipment (UE) receives one or more timing advance commands from a non-terrestrial network (NTN). The UE resets, after a GNSS fix, a cumulative timing advance value based on the one or more timing advance commands from the NTN and transmits an uplink transmission with a timing advance based on a self-estimated delay and the cumulative timing advance value. The UE may adjust a self-estimated delay based on at least one GNSS fix. The UE may transmit an uplink transmission with a timing advance based on the adjusted self-estimated delay and a cumulative timing advance value based on the one or more timing advance commands from the NTN.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may determine a synchronization signal/physical broadcast channel (SS/PBCH) block index, associated with an SS/PBCH block, based at least in part on a set of bits included in a payload of a PBCH of the SS/PBCH block, a demodulation reference signal (DMRS) sequence included in the PBCH, a scrambling sequence associated with the PBCH, or an interleaving associated with the PBCH. Numerous other aspects are described.

Abstract: Certain aspects of the present disclosure provide techniques that may allow a base station and user equipment (UE) to be in agreement on spatial quasi co-location (QCL) assumptions for transmission in certain scenarios. The techniques may be used, for example, to determine QCL assumptions for downlink transmissions sent in an initial control resource set (CORESET) on an initial downlink bandwidth part (BWP) where dedicated signaling of such assumptions may be lacking. In some cases, the techniques may involve a UE determining a beam to use for monitoring for a physical downlink control channel (PDCCH) transmission on the initial CORESET in the initial downlink BWP that lacks dedicated signaling of QCL information and monitoring for the PDCCH on the initial CORESET in the initial BWP using the determined beam. Other aspects and features are also claimed and described.

Abstract: This application discloses a packet sending method, apparatus, and system, and a storage medium, and relates to the communication field. The method includes: A first device receives a first packet, where the first packet includes a first address prefix, the first device is a border device in a first AS domain, the first address prefix is obtained based on an address prefix advertised by a second device, the second device is located in a second AS domain, the first AS domain is different from the second AS domain, and the first address prefix is used to send a packet to the second device. The first device obtains a second address prefix and identification information of a target flexible algorithm based on the first address prefix, where the second address prefix is used to send a packet to the second device.