Abstract: The system can include a set of joints, a controller, and a model engine; and can optionally include a support structure and an end effector. Joints can include: a motor, a transmission mechanism, an input sensor, and an output sensor. The system can enable articulation of the plurality of joints.

Abstract: A controller for executing a task based on probabilistic image-based landmark localization, uses a neural network, which is trained to process images of objects of a type having a structured set of landmarks to produce a parametric probability distribution defined by values of parameters for a location of each landmark in each processed image. The controller submits the set of input images to the neural network to produce the values of the parameters that define the parametric probability distribution over the location of each landmark in the structured set of landmarks of each input image. Further, the controller determines, for each input image, a global landmark uncertainty for the image based on the parametric probability distributions of landmarks in the input image and executes the task based on the parametric probability distributions of landmarks in each input image and the global landmark uncertainty of each input image.

Abstract: Methods, systems, and non-transitory computer readable medium are described for automated image measurement for process development and optimization. An example method may include receiving an image of a product associated with a manufacturing process, wherein the product comprises a plurality of structures; identifying, using a trained machine learning model, a segment of the image that comprises a structure of the plurality of structures; determining a plurality of image measurements of the segment that comprises the structure; and storing the plurality of image measurements.

Abstract: Systems and methods in accordance with embodiments of the invention implement all-microwave stabilized microresonator-based optical frequency comb. In one embodiment, an all-microwave stabilized microresonator-based optical frequency comb includes: an optical pump configured to generate pulses of light; a microresonator including an input configured to receive pulses generated by an optical pump and an output configured to generate an optical frequency comb signal characterized by frep and ?; where frep describes spacing of frequency components in the optical frequency comb; where the optical frequency comb includes a primary comb and a plurality of subcombs and ? is a frequency offset between subcombs; and two phase locked loops that phase lock frep and ? to low noise microwave oscillators by modulating output power and pump frequency of the optical pump.

Abstract: In implementations of reviewing document designs, a document review system can import reviewer comments to a design application used to author a document design. The reviewer comments can be made by a reviewer via a review application implemented on an additional computing device other than the computing device implementing the design application. The document review system can add, via the design application, a reply comment to the reviewer comment. The reply comment can be made by a designer of the document design as part of a comment hierarchy that indicates comment sequences and comment links for the document design. The document review system can also export, to the review application, a review document representing the document design that preserves the comment hierarchy in the document design.

Abstract: Based on graphene heterostructure in chip-scale silicon nitride microresonators, optoelectronic control and modulation in frequency combs via group velocity dispersion modulation can be demonstrated. By tuning graphene Fermi level from 0.50 eV to 0.65 eV via electric-field gating, deterministic in-cavity group velocity dispersion control from anomalous (?62 fs2/mm) to normal (+9 fs2/mm) can be achieved with Q factor remaining high at 106. Consequently, both the primary comb lines and the full comb spectra can be controllable dynamically with the on/off switching of the Cherenkov radiation, the tuning of the primary comb lines from 2.3 THz to 7.2 THz, and the comb span control from zero comb lines to ˜781 phase-locked comb lines, directly via the DC voltage.

Abstract: A method includes determining a plurality of features of a first original image of a first product that are expected to be different for one or more products to be produced via manufacturing parameters of a manufacturing process compared to the first product. The method further includes adjusting one or more of the plurality of features of the first original image to generate a first synthetic image. The method further includes providing a plurality of images including the first original image and the first synthetic image to train a machine learning model to generate a trained machine learning model configured to generate output associated with updating the manufacturing parameters of the manufacturing process.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for dialogue systems. A transcription of a user utterance is obtained. The transcription of the utterance is tokenized to identify multiple tokens for the utterance. Token-level utterance encodings corresponding to different tokens of the transcription are generated. A system action encoding from data indicating system actions previously performed by the dialogue system are generated. A dialogue context vector based on the utterance encoding and the system action encoding are generated. The token-level utterance encodings, the system action encoding, and the dialogue context vector are processed using a slot tagger to produce token-level output vectors. A limited set of candidate token classifications for the tokens of the user utterance are determined based on the token-level utterance encodings. A response for output is provided in response to the user utterance.

Abstract: A method includes providing attributes of a manufacturing process and an image of a product associated with the manufacturing process to a trained machine learning model. The method further includes obtaining, from the trained machine learning model, predictive data. The method further includes determining, based on the predictive data, image measurements of the image of the product associated with the manufacturing process. Manufacturing parameters of the manufacturing process are to be updated based on the image measurements.

Abstract: Generally, the present disclosure is directed to systems and methods for performing task-oriented response generation that can provide advantages for artificial intelligence systems or other computing systems that include natural language processing for interpreting user input. Example implementations can process natural language descriptions of various services that can be accessed by the system. In response to a natural language input, systems can identify relevant values for executing one of the service(s), based in part on comparing embedded representations of the natural language input and the natural language description using a machine learned model.

Abstract: A controller for executing a task based on probabilistic image-based landmark localization, uses a neural network, which is trained to process images of objects of a type having a structured set of landmarks to produce a parametric probability distribution defined by values of parameters for a location of each landmark in each processed image. The controller submits the set of input images to the neural network to produce the values of the parameters that define the parametric probability distribution over the location of each landmark in the structured set of landmarks of each input image. Further, the controller determines, for each input image, a global landmark uncertainty for the image based on the parametric probability distributions of landmarks in the input image and executes the task based on the parametric probability distributions of landmarks in each input image and the global landmark uncertainty of each input image.

Abstract: The present disclosure involves systems, software, and computer implemented methods for providing voice-enabled human tasks in process modeling. One example method includes receiving a deployment request for a workflow that includes a human task. The workflow is deployed to a workflow engine in response to the deployment request. An instance of the workflow is created in response to a request from a client application. The instance of the workflow is processed, including execution of the human task. The human task is added to a task inbox of an assignee of the human task. A request is received from the assignee to access the task inbox from a telecommunications system. Voice guidance is provided, to the assignee, that requests assignee input. Voice input from the assignee is processed for completion of the human task. Workflow context for the human task is updated based on the received voice input.

Abstract: Methods, systems, and non-transitory computer readable medium are described for automated image measurement for process development and optimization. A method includes receiving original images including a first original image. Each original image is of a corresponding product associated with a manufacturing process. The method further includes performing, based on process information about the manufacturing process, feature extraction to identify features of the first original image that are expected to change based on manufacturing parameters of the manufacturing process. The method further includes generating a first synthetic image of synthetic images by performing targeted deformation of one or more of the features of the first original image.

Abstract: Methods, systems, and non-transitory computer readable medium are described for automated image measurement for process development and optimization. A method includes receiving an image of a product associated with a manufacturing process; determining, using a trained machine learning model, an image classification for the image; selecting, based on the image classification, one or more image processing algorithms for the image; pre-processing the image based on at least one of the one or more image processing algorithms to generate an enhanced image; measuring, using a first image processing algorithm of the one or more image processing algorithms, one or more attributes of the enhanced image to determine image measurements; and reporting the image measurements. The manufacturing parameters of the manufacturing process are to be updated based on the image measurements.

Abstract: In implementations of reviewing document designs, a document review system can import reviewer comments to a design application used to author a document design. The reviewer comments can be made by a reviewer via a review application implemented on an additional computing device other than the computing device implementing the design application. The document review system can add, via the design application, a reply comment to the reviewer comment. The reply comment can be made by a designer of the document design as part of a comment hierarchy that indicates comment sequences and comment links for the document design. The document review system can also export, to the review application, a review document representing the document design that preserves the comment hierarchy in the document design.

Abstract: Based on graphene heterostructure in chip-scale silicon nitride microresonators, optoelectronic control and modulation in frequency combs via group velocity dispersion modulation can be demonstrated. By tuning graphene Fermi level from 0.50 eV to 0.65 eV via electric-field gating, deterministic in-cavity group velocity dispersion control from anomalous (?62 fs2/mm) to normal (+9 fs2/mm) can be achieved with Q factor remaining high at 106. Consequently, both the primary comb lines and the full comb spectra can be controllable dynamically with the on/off switching of the Cherenkov radiation, the tuning of the primary comb lines from 2.3 THz to 7.2 THz, and the comb span control from zero comb lines to ˜781 phase-locked comb lines, directly via the DC voltage.

Abstract: A method for managing Logical Block Address (LBA) range overlap checking in a Non-Volatile Memory express (NVMe) based Solid State Drive (SSD) includes detecting, by an LBA-Overlap Check (LOC) module, an overlap between an LBA range of an incoming command with an LBA range of at least one outstanding command in an SSD controller, determining, by the LOC module, an overlap count value corresponding to the incoming command, where the overlap count value indicates occurrence of an overlap between the LBA range of the incoming command and the LBA range of the at least one outstanding command, and executing, by the SSD controller, the incoming command based on the overlap count value corresponding to the incoming command.

Abstract: A method for managing Logical Block Address (LBA) range overlap checking in a Non-Volatile Memory express (NVMe) based Solid State Drive (SSD) includes detecting, by an LBA-Overlap Check (LOC) module, an overlap between an LBA range of an incoming command with an LBA range of at least one outstanding command in an SSD controller, determining, by the LOC module, an overlap count value corresponding to the incoming command, where the overlap count value indicates occurrence of an overlap between the LBA range of the incoming command and the LBA range of the at least one outstanding command, and executing, by the SSD controller, the incoming command based on the overlap count value corresponding to the incoming command.

Abstract: A method for handling a command ID conflict in an NVMe-based solid-state drive (SSD) device includes fetching, from a host submission queue (HSQ), one or more commands submitted by a host device. The fetched commands are checked to determine if there is a command ID conflict. A command ID (CID) error interrupt is communicated to firmware of the SSD device if the command ID conflict is detected. A command validation is performed for the one or more commands on receiving the CID error interrupts. A command response is communicated with additional special information from the device FW to the host device for a command having a command ID conflict. One or more resources associated with the one or more commands are released based on the command response.

Abstract: Methods, systems, and non-transitory computer readable medium are described for automated image measurement for process development and optimization. A method includes receiving original images including a first original image. Each original image is of a corresponding product associated with a manufacturing process. The method further includes performing, based on process information about the manufacturing process, feature extraction to identify features of the first original image that are expected to change based on manufacturing parameters of the manufacturing process. The method further includes generating a first synthetic image of synthetic images by performing targeted deformation of one or more of the features of the first original image.