Abstract: A method, system, and non-transitory computer-readable medium for characterizing oxide defects in semiconductor devices are described. Based on time-resolved SILC data, a total number of active oxide defects and each defect's characteristics can be determined for a deeply-scaled FET device. The discrete changes in leakage current correspond to switching of a single defect. A Bayesian-inspired algorithm is utilized to extract distinct current levels and experimental data is quantized into these extracted current levels by filtering noise. The evolution of current levels corresponds to a Markov chain. The defect currents are extracted by clustering transition probabilities and absolute differences in current levels using an affinity propagation algorithm. A maximum likelihood estimator is developed to extract a base leakage current.

Abstract: A system includes at least one processing circuit including at least one memory and one or more processors configured to: obtain, via a user interface, a non-bounded query relating to a travel trip, the non-bounded query excluding at least one parameter associated with the travel trip; identify, via a first machine learning model, the at least one excluded parameter based on the non-bounded query; identify, via a second machine learning model and from among a first set of travel results that match the non-bounded query and the excluded parameter, a second set of the travel results that meet a presentation threshold of customer popularity based on at least one of a location selected by one or more users or a time selected by the one or more users; and present, via the user interface, one or more travel results of the second set, and the presentation threshold.

Abstract: In implementations of systems for generating indications of relationships between electronic documents, a processing device implements a relationship system to segment text of electronic documents included in a document corpus into segments. The relationship system determines a subset of the electronic documents that includes electronic document pairs having a number of similar segments that is greater than a threshold number. The similar segments are identified using locality sensitive hashing. The electronic document pairs are classified as related documents or unrelated documents using a machine learning model that receives a pair of electronic documents as an input and generates an indication of a classification for the pair of electronic documents as an output. Indications of relationships between particular electronic documents included in the subset are generated based at least partially on the electronic document pairs that are classified as related documents.

Abstract: Various embodiments of the present technology may provide a susceptor and a sensing system coupled to the susceptor and configured to generate a plurality of sensor output signals indicating air flow across the susceptor and sensing system. A controller is connected to the sensing system and configured to detect a flow pattern on the susceptor based on the sensor output signals.

Abstract: Described herein are systems and techniques to facilitate the efficient performance of operations and determination of operation states and related data. A set of operations and operation dependencies may be configured at an interactive nonlinear operation set navigator that may generate interface elements representing the operations and operation data. Operation interface elements may further include controls that, when activated, generate additional operation data and/or facilitate the initiation and completion of the represented operation. The duration and other properties associated with the execution of an operation may be tracked and used to generate alerts and perform other actions.

Abstract: A method including, at an autonomous vehicle: autonomously navigating across an outdoor terrain to locate a robotic arm and a set of objects proximal an install location; accessing an image from an optical sensor; detecting a set of install features at the install location based on the image; calculating a gross install pose of the object that locates an object proximal the install location and offset from the set of install features; defining a keep-in boundary proximal the install location and encompassing the gross install pose; autonomously navigating the robotic arm to retrieve the object from the set of objects and to locate the object in the gross install pose; detecting a series of forces applied to a distal end of the robotic arm; and navigating the object in directions of the series of forces while maintaining the object fully within the keep-in boundary.

Abstract: Techniques are described for a machine learning system configured to generate respective sample embeddings for a plurality of sample statements. The machine learning system may further be configured to generate a statement embedding for a statement. The machine learning system may further be configured to determine, based on the sample embedding and the statement embedding, respective similarity scores for the sample embeddings. The machine learning system may further be configured to select, based on the respective similarity scores for the sample embeddings, one or more sample statements from the plurality of sample statements. The machine learning system may further be configured to generate a prompt including the one or more sample statements, the statement, and at least one of respective ground-truth information or respective paraphrases for the selected one or more sample statements. The machine learning system may further be configured to provide the prompt to a machine learning model.

Abstract: Malformed VLAN packets can be detected by programming suitable rules in a TCAM in the packet processing pipeline. In some deployments, for example, the TCAM rule(s) can match on the parsed EtherType metadata. More specifically, the match can be based on the EtherType metadata being set to a value equal to known VLAN TPIDs, such as 0x8100, 0x88a8, rather than being set to a standard EtherType.

Abstract: Systems and methods that support the creation and timely electronic scheduling and delivery of course materials for training of individuals in an organization, in which course recommendations, scheduling, and rollout are based upon a number of factors including, for example, specific individual and/or business day-to-day operational performance measures, sales performance, and seasonal weather conditions by geographical region.

Abstract: A signal sampling circuitry comprises: a plurality of sampling units receiving an input signal for time-interleaved sampling, each sampling unit comprising: a sampling capacitor having a first plate connected to an output of the sampling unit; a first plate switch between the first plate and a first reference voltage, a second plate switch between a second plate of the sampling capacitor and a second reference voltage; an input buffer for outputting a buffered input signal to the second plate; wherein the input buffer is connected to at least one power gating switch for powering down the input buffer.

Abstract: . There is great interest in developing strategies to identify proteolytic substrates. Synthetic peptide libraries are among the most widely used to screen substrates-based activity, however, this technique is limited by the size of libraries and is time-consuming. Several combinatorial library-based display technologies such as phage, bacteria and yeast display, can identify better substrates for protease activity with higher-throughput sequencing and enrichment of specific substrates over multiple rounds of selection. Even though these profiling techniques provide greater coverage of protease cleavable substrates, such cells are less capable of accessing difficult to deliver sites, pose immunogenicity concerns, and have not been applied to discriminate on-from off-target activity in vivo. Disclosed herein are in vitro and in vivo methods for screening protease substrates.

Abstract: Systems and methods related to priority-based and performance-based selection of a render mode, such as a two-level binning mode, in which to execute workloads with a graphics processing unit (GPU) of a system are provided. A user mode driver (UMD) or kernel mode driver (KMD) executed at a central processing unit (CPU) configures low and medium priority workloads to be executed in a two-level binning mode and selects a binning mode for high priority workloads based on whether performance heuristics indicate that one or more binning conditions or override conditions have been met. High priority workloads are maintained in a high priority queue, while low and medium priority workloads are maintained in a low/medium priority queue, such that execution of low and medium priority workloads at the GPU can be preempted in favor of executing high priority workloads.

Abstract: Systems, devices, media, and methods are presented for device independent secure messaging. The systems and methods generate an encrypted message by encrypting message content, designated for a specified recipient, with an encryption key. The systems and methods select a set of recipient keys, associated with the specified recipient, from a plurality of member keys. For each recipient key, the systems and methods encrypt the encryption key to generate a set of encrypted keys and transmit the encrypted message and an encrypted key of the set of encrypted keys to one or more client devices associated with the specified recipient. The systems and methods then receive an acknowledgement indicating a termination status of the encrypted message.

Abstract: A method and computing apparatus for extracting information from a document are provided. The method includes receiving a document, extracting data from the document, assigning the document to a category from among a predetermined plurality of categories based on a result of the extracted data, and generating a structured output by formatting the extracted data based on the assigned category.

Abstract: A global count or reference time in a computing device may be maintained during a sleep state in which the global counter is powered off. The global count from the global counter may be saved in a register when the sleep state is entered. When the sleep state is exited, the global count in the register may be restored in the global counter.

Abstract: Systems, devices, and methods provide for smart contract automation for triggering smart contract functions. The smart contract automation may receive log data, provide an upkeep payload to a workflow, the upkeep payload comprising one or more active upkeeps. The smart contract automation may determine, by the proposal workflow, whether the one or more active upkeeps is an eligible upkeep. The smart contract automation may determine, by the smart contract automation system, a reportable result based on the eligible upkeep and the log data. The smart contract automation may transmit the reportable result to a blockchain.

Abstract: Systems and methods are provided for distributed video storage and search over edge computing devices having a short-term memory and a long-term memory. The method may comprise caching a first portion of data on a first device. The method may further comprise determining, at a second device, whether the first device has the first portion of data. The determining may be based on whether the first piece of data satisfies a specified criterion. The method may further comprise sending the data, or a portion of the data, and/or a representation of the data from the first device to a third device.

Abstract: A reference buffer circuit for buffering a reference voltage is disclosed. The reference buffer circuit includes an input circuit that may generate a replica of the reference voltage. A feedback circuit may generate a charging current using a current mirror circuit and a bias current. A value of the charging current may be greater than a value of the bias current. The feedback circuit may provide a buffered version of the reference voltage at a reference node using the charging current and the replica voltage. An output stage circuit may sink a compensation current from the reference node.

Abstract: A data loss prevention (DLP) pipeline is presently disclosed filters non-sensitive documents from DLP with two stages—a first filtering stage that filters documents that do not match one or more patterns corresponding to personally identifiable information (PII) and a second filtering stage that filters documents matching the one or more patterns that are classified with non-sensitive verdicts by a machine learning (ML) ensemble. The ML ensemble comprises a character-level convolutional neural network that generates pattern-based embeddings, a language model that generates context-based embeddings, and a classification model that receives concatenated embeddings as input to generate sensitive/non-sensitive verdicts.