Abstract: Methods, systems, apparatuses, devices, and computer program products are described. A flow generation service may receive a natural language input that indicates instructions for automating a task according to a first process flow. Using a large language model (LLM), the flow generation service may decompose the natural language input into a set of elements (e.g., logical actions) and connectors, where the LLM may be trained on first metadata corresponding to a second process flow that is created manually by a user. In addition, using the LLM, the flow generation service may generate second metadata corresponding to each of the set of elements based on decomposing the natural language input. The flow generation service may sequence and merge the set of elements to generate the first process flow. In some examples, the flow generation service may send, for display to a user interface of a user device, the first process flow.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment may receive a configuration for monitoring one or more physical downlink control channels (PDCCHs) for downlink control information (DCI) that schedules physical downlink shared channel (PDSCH) communications on each carrier included in a set of carriers and for DCI that schedules physical uplink shared channel (PUSCH) communications on one or more carriers. The UE may monitor the one or more PDCCHs for DCI in accordance with the configuration. The UE may receive DCI that schedules a PUSCH communication on the one or more carriers based at least in part on the monitoring. Numerous other aspects are provided.

Abstract: Embodiments described herein provide a method of predicting an action by a plurality of language model augmented agents (LAAs). In at least one embodiment, a controller receives a task instruction to be performed using an environment. The controller receives an observation of a first state from the environment. The controller selects a LAA from the plurality of LAAs based on the task instruction and the observation. The controller obtains an output from the selected LAA generated using an input combining the task instruction, the observation, and an LAA-specific prompt template. The controller determines the action based on the output. The controller causes the action to be performed on the environment thereby causing the first state of the environment to change to a second state.

Abstract: A network device sends downlink control information to a first terminal device. The downlink control information includes: time domain resource information, where the time domain resource is a time domain resource on which a first data channel is located; and frequency domain resource information, where the frequency domain resources are a frequency domain resource on which the first data channel is located and a frequency domain resource on which a second data channel is located. The second data channel is located on a time domain resource different from the time domain resource on which the first data channel is located. The first terminal device receives, on the determined time domain resource and frequency domain resource, a signal that is located on the first data channel and that is from the network device.

Abstract: Embodiments described herein provide for optimizing a language model (LM) agent. In at least one embodiment, and LM agent comprises an “actor” LM and a “retrospective LM which provides reflections on attempts by the actor LM. The reflections are used to update subsequent prompts to the actor LM. Optimizing the LM agent comprises fine-tuning parameters of the retrospective LM while keeping parameters of the actor LM frozen. A gradient may be determined by a change in reward from the environment based on actions taken by the actor LM with and without a reflection of the retrospective LM. Using this gradient, parameters of the retrospective LM may be updated via backpropagation.

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: The present disclosure is related to systems and methods for processing X-ray images. A method for processing an X-ray image may include obtaining an X-ray image; and determining a metal image based on the X-ray image by using a trained metal detection model. The metal image includes information of a metal object in the X-ray image.

Abstract: A method for patterning a stack having a mask with a plurality of mask features is provided. A targeted deposition is provided, wherein the targeted deposition comprises a plurality of cycles, wherein each cycle comprises flowing a precursor to deposit a layer of precursor and targeted curing the layer of precursor, comprising flowing a curing gas, flowing a modification gas, forming a plasma from the curing gas and modification gas, and exposing the layer of precursor to the plasma providing a targeted curing, wherein plasma from the curing gas cures first portions of the layer of precursor and plasma from the modification gas modifies second portions of the layer of precursor, wherein the modification of the second portion reduces curing of the layer of precursor of the second portions of the layer of precursor. The stack is etched through the targeted deposition.

Abstract: The present disclosure relates to systems and methods for image processing. The system may obtain at least one image of an object. For each of the at least one image, the system may determine a recognition result of the image. The recognition result may include an image type of the image, a type of a lesion in the image, a region of the lesion in the image, and/or an image feature of the image. Further, the system may process the at least one image of the object based on at least one recognition result corresponding to the at least one image.

Abstract: Embodiments of a method performed in a User Plane Entity (UPE) in a core network to perform a user plane initiated Packet Forwarding Control Protocol (PFCP) association release procedure are provided. In some embodiments, the method comprises, for each affected PFCP session, sending, to a Control Plane Entity (CPE), a PFCP session report request comprising an indicator that the PFCP session is being removed and one or more usage reports for the PFCP session, and deleting the PFCP session. The method further comprises sending, to the CPE, a PFCP association update request comprising information that indicates that the PFCP association update request is to initiate release of the PFCP association and an indicator that all non-zero usage reports for the affected PFCP session(s) have been reported. The method further comprises receiving, from the CPE, a PFCP association release request and sending, to the CPE, a PFCP association release response.

Abstract: Wireless communications systems and methods related to communicating a serving cell dormancy indication field and acknowledgement. In some implementations, a wireless communication device (e.g., user equipment), may detect a physical downlink control channel (PDCCH) with a secondary cell (Scell) dormancy indication field. The user equipment can change states from an initial state to another state (e.g., a dormancy-like state to a non-dormancy like state) based on detecting Scell indicator field in the PDCCH. The user equipment may also transmit a hybrid-ARQ acknowledgment (HARQ-ACK) in response to the detecting the PDCCH with the Scell dormancy indication field. Other aspects and features are also claimed and described.

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: 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: Disclosed are preparation for an ethinylestradiol-cyclodextrin complex and use thereof. The ethinylestradiol-cyclodextrin complex utilizes the properties of cyclodextrin that the inner cavity is hydrophobic and the outer surface is hydrophilic to increase the water solubility of the ethinylestradiol-cyclodextrin complex that is also an ethinylestradiol drug, so that the solubility of ethinylestradiol is greatly increased and the bioavailability of water-insoluble ethinylestradiol is improved, thereby providing the possibility for changing the administration mode and homogeneous injection of ethinylestradiol.

Abstract: Embodiments of the present specification disclose a system and method for generating a vehicle damage image on the basis of a GAN model. During operation, the system obtains a real vehicle image, generates an intermediate image based on the real vehicle image by labeling a target box on the real vehicle image and removing a portion of the real vehicle image within the target box, and generates the vehicle damage image based on the intermediate image by inputting the intermediate image into a machine-learning model, which outputs the vehicle damage image by filling a local image indicating vehicle damage into the target box of the intermediate image.

Abstract: The present disclosure relates to the technical field of genetic engineering, in particular to a target ligand. The target ligand provided by the present disclosure forms a siRNA conjugate with a specific small interfering RNA sequence, which targets to an angiopoietin like 3 (ANGPTL3), and reduce the expression quantity of an ANGPTL3 protein by degrading a transcript of an ANGPTL3 gene in a cell. Therefore, the siRNA conjugate formed by the target ligand provided by the present disclosure may be used to prevent and/or treat a dyslipidemia disease.

Abstract: The present invention belongs to the technical field of subsea equipment, and specifically pertains to a thrust reversal type emergency jettison device for an underwater vehicle. A thrust reversal type emergency jettison device for an underwater vehicle, comprises a thrust reversal mechanism, the thrust reversal mechanism comprises 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.

Abstract: This patent discloses a medical service platform, method, and apparatus based on a human digital twin model that comprises a a data currency valuation subsystem that provides dynamic quotes of a digital currency based on values of corresponding data in the digital human digital data currency system.

Abstract: This disclosure provides a medical and healthcare service platform that is supported by a digital data currency system and provides medical and healthcare data processing, analyzing, and predicting based on a digital human system by integrating participating parties comprising individual persons, researchers, healthcare providers, and regulatory and public sectors.

Abstract: An embodiment provides a system and method for sample labeling. During operation, the system obtains a plurality of historical loss assessment images and obtains a plurality of candidate samples from the plurality of loss assessment images. A respective candidate sample comprises an image of a candidate damage area detected in a corresponding historical loss assessment image. The system clusters the plurality of candidate samples into a plurality of class clusters. For a respective class cluster, the system determines a center candidate sample set corresponding to a class cluster center of the respective class cluster, receives a manual labeling result associated with candidate samples in the determined center candidate sample set, and performs, according to the manual labeling result, damage classification labeling on other unlabeled candidate samples in the respective class cluster to obtain a plurality of labeled samples.