Abstract: Various embodiments discussed herein relate to prompting a model, such as a Large Language Model (LLM), to ingest natural language clustering instructions and generate corresponding natural language clustering information, such as a cluster description and/or a cluster label without the need to generate any numeric text embeddings.

Abstract: A database-aware storage server provides instant creation of snapshots without the need to create an intermediate test master database. During the snapshot creation time, the source database stays read-write and completes ongoing reads and writes. The database-aware storage server allows creation of layers of writable snapshots in a hierarchy. All these multiple databases share common data blocks. Any new writes performed by the database post snapshot are stored in blocks of sparse files. This promotes space sharing and reduces the total amount of space used by all these related databases. The allocations for the source and all new snapshot databases share the same common pool of storage. The newly created snapshot databases can access the data store directly without going through an intermediate layer.

Abstract: A package securing system operable to facilitate the ability for an individual expecting to receive a package to provide and control a secure environment for receiving the package wherein the individual can execute the security protocols remotely. The package securing system further includes a lock having a body, a shank with keeper operably secured thereto, a solenoid with piston, a scanner and a controller. The controller provides operation of the lock and is operably coupled to the Internet via a wireless communication protocol. The controller is communicably coupled to a software application wherein the software application provide an interface for operation of the package securing system. The lock is unlocked and locked utilizing a machine-readable dataset wherein the machine-readable dataset is associated with an expected package and communicated to the controller via the software application. A server is provided to execute portions of the operational method of the present invention.

Abstract: Embodiments of the disclosure are in the field of advanced integrated circuit structure fabrication and, in particular, integrated circuit structures having germanium-based channels are described. In an example, an integrated circuit structure includes a fin having a lower silicon portion, an intermediate germanium portion on the lower silicon portion, and a silicon germanium portion on the intermediate germanium portion. An isolation structure is along sidewalls of the lower silicon portion of the fin. A gate stack is over a top of and along sidewalls of an upper portion of the fin and on a top surface of the isolation structure. A first source or drain structure is at a first side of the gate stack. A second source or drain structure is at a second side of the gate stack.

Abstract: Disclosed herein are etching methods, patterned substrates made using said methods, and methods of use of said patterned substrates. For example, described herein are methods comprising: oxidizing at least a portion of a surface of a substrate comprising a group III-V compound, thereby forming an oxidized layer at said portion of the surface; and etching the oxidized layer by exposing the oxidized layer to an etchant; wherein: the group III-V compound comprises one or more group III elements (e.g., Al, Ga, In, B, Sc, Y, or a combination thereof) and a group V element (e.g., N, As, or Sb); the etchant comprises at least one of the group III elements; and the etchant removes oxides by the etchant reacting with the oxide to form a suboxide, which desorbs from the surface.

Abstract: In some examples, a data management and storage (DMS) platform comprises peer DMS nodes in a node cluster, a distributed data store comprising local and cloud storage, and at least one processor configured to perform operations in a method of creating a local consolidated patch file from a patch file chain stored in the cloud storage. Example operations comprise, in a first dry-run phase, creating a patch file image of data blocks in one or more cloud patch files stored in the cloud storage; in a second data-transfer phase, downloading at least some of the data blocks from the cloud patch files identified by the patch file image; and creating and storing, in the local storage, the local consolidated patch file using the downloaded data blocks.

Abstract: Methods, computer systems, computer-storage media, and graphical user interfaces are provided for efficiently generating and using intent taxonomies. In embodiments, training data, including data requests for information, is obtained. Thereafter, a model prompt to be input into a large language model is generated. The model prompt includes an instruction to generate an intent taxonomy, an indication of the training data to use for generating the intent taxonomy, and a taxonomy attribute desired to be used as criteria to generate a quality intent taxonomy. An intent taxonomy that includes user intent classes is obtained as output from the large language model. The intent taxonomy is analyzed to determine whether the intent taxonomy is valid. When the intent taxonomy is determined as valid, the intent taxonomy is provided for use in identifying user intent, and when the intent taxonomy is determined as invalid, the intent taxonomy is refined.

Abstract: Proactive caching, at a client device (e.g., a feature phone), of transient assistant action suggestions for selective rendering by an assistant client application of the client device. A transient assistant action suggestion, when rendered via an assistant client application and selected, causes the assistant client application to initiate performance of a corresponding assistant action. In various implementations, a prefetched transient action suggestion can be a time-constrained suggestion that includes at least associated rendering restriction metadata that defines one or more temporal windows to which rendering of the time-constrained suggestion is restricted. Proactive cache refresh rate metadata can also be associated with transient action suggestion(s) and defines a duration during which the assistant client application is to refrain from interfacing with a remote system to prefetch updated transient assistant action suggestions.

Abstract: Techniques are disclosed for monitoring a patient for the occurrence of a cardiac arrhythmia. A computing system generates sample probability values by applying a machine learning model to sample patient data. The machine learning model determines a respective probability value that indicates a probability that the cardiac arrhythmia occurred during each respective temporal window. The computing system outputs a user interface comprising graphical data based on the sample probability values and receives, via the user interface, an indication of user input to select a probability threshold for a patient. The computing system receives patient data for the patient and applies the machine learning model to the patient data to determine a current probability value. In response to the determination that the current probability exceeds the probability threshold for the patient, the computing system generates an alert indicating the patient has likely experienced the occurrence of the cardiac arrhythmia.

Abstract: Front-end circuitry is difficult to design for high sample rate, wide bandwidth receivers with high performance requirements on noise and linearity. One exemplary front-end circuitry is integrated on-chip with the RF ADC in a receiver, and the circuitry implements ESD protection, attenuation, and gain. The circuitry includes a multi-tap filter with LC circuits, and the filter implements a highly linear filter. Advantageously, the capacitors in the LC circuits are also used for ESD protection. Additionally, tunable attenuator cells are implemented across the multi-tap filter to provide a wide range of variable attenuation. The circuitry can further include a fixed or variable gain stage at the output. The resulting circuitry offers variable gain and attenuation while meeting bandwidth, noise, and linearity requirements.

Abstract: A foetal monitoring device for simultaneously monitoring a concentration of an analyte in a foetus and a heart rate of the foetus, the device comprising a biosensor and an electrocardiogram (ECG).

Abstract: In some examples, a method for developing an application programming interface (API) is provided. The method is performed by a computing system running an API Forge Accelerator. The method comprising: receiving, by the API Forge Accelerator, a request for generating an API for a service based on a type of a first gateway, wherein the request indicates a selection of the first gateway among a plurality of gateways integrated with the API Forge Accelerator; generating, by the API Forge Accelerator, based on the specific gateway, a specification template corresponding to the specific gateway for generating the API, wherein the specification template indicates specification information that is used for the API generation; obtaining, by the API Forge Accelerator, based on the specification template, specification information from user input and API tools; and generating, by the API Forge Accelerator, based on the specification information, the API for the service.

Abstract: A system and method for encrypting portions of data for storage in a remote network have been provided. The system comprises a memory with instructions executable by a processor to receive data for forwarding to a server device, wherein the received data comprises an indication of one or more portions of the received data to be encrypted; identify a portion comprising the one or more portions of the received data based at least in part on the indication; encrypt the identified portion of the data; generate a payload that comprises the encrypted portion and one or more unencrypted portions of the received data; and transmit, to the server device, the payload.

Abstract: Techniques are disclosed for monitoring a patient for the occurrence of cardiac arrhythmias. A computing system obtains a cardiac electrogram (EGM) strip for a current patient. Additionally, the computing system may apply a first cardiac rhythm classifier (CRC) with a segment of the cardiac EGM strip as input. The first CRC is trained on training cardiac EGM strips from a first population. The first CRC generates first data regarding an aspect of a cardiac rhythm of the current patient. The computing system may also apply a second CRC with the segment of the cardiac EGM strip as input. The second CRC is trained on training cardiac EGM strips from a smaller, second population. The second CRC generates second data regarding the aspect of the cardiac rhythm of the current patient. The computing system may generate output data based on the first and/or second data.

Abstract: Disclosed herein is a method and apparatus for synthesizing co-crystals from the vapor phase without the need for liquid solvent. Also disclosed herein are co-crystals formed from the vapor phase, substrates coated with said co-crystals, pharmaceutical compositions thereof and apparatuses for producing said co-crystals.

Abstract: According to one embodiment, a method, computer system, and computer program product for coordinating professional language is provided. The present invention may include running, by a processor, at least one professional parsing algorithm and at least one non-professional parsing algorithm on an ingested writing; assigning, by at least one of the professional parsing algorithms and at least one of the non-professional parsing algorithms, one or more initial scores to the ingested writing; determining, by a machine learning model, one or more differences between the ingested writing and stored professional data by comparing the one or more initial scores of the ingested writing to the stored professional data; determining a variation percentage score of the ingested writing based on the differences; and transmitting at least one suggested modification of the ingested writing based on the variation percentage score to a transmitting device.

Abstract: A cordless endoscope assembly includes a housing, an endoscope tube and an eyepiece. Electronics are housed in the housing and operable to operate one or more light sources in the endoscope, the electronics including one or more batteries and a printed circuit board. The housing includes a connector configured to removably couple with a cannula connector of a cannula when the endoscope tube is inserted through the cannula. The cordless endoscope is a disposable single use endoscope that can optionally be sterilized with an ethylene oxide (EtO) sterilization process.

Abstract: This specification describes a computer-implemented method of training a machine-learned speech audio generation system for use in video games. The training comprises: receiving one or more training examples. Each training example comprises: (i) ground-truth acoustic features for speech audio, (ii) speech content data representing speech content of the speech audio, and (iii) a ground-truth speaker identifier for a speaker of the speech audio.

Abstract: A system provides for locking or unlocking a lock at a door of an intelligent door lock system at a dwelling by a designated dwelling user. An intelligent door lock system includes a remotely operable lock at a dwelling accessible by a user and a server in communication with the intelligent door lock system. A user mobile device is in communication with the server. The user mobile device includes an App that provides for locking and/or unlocking of the lock of the intelligent door lock system. The user communicates with the server using the user's mobile device, and the server is configured to process the user crossings one or more geo-locations relative to the dwelling and create a data set that is stored in a database.

Abstract: A method may include, when in a low-power mode of a power converter, monitoring an output voltage of the power converter; comparing, with a single comparator, the output voltage to a first threshold voltage and cause the power converter to enter a magnetization phase of the low-power mode responsive to the output voltage falling below the first threshold voltage; and during the magnetization phase and a demagnetization phase of the low-power mode, comparing, with the single comparator, the output voltage to a second threshold voltage lower than the first threshold voltage and cause the power converter to enter the high-power mode responsive to the output voltage falling below the second threshold voltage.