Abstract: Example embodiments of aspects of the present disclosure provide an example computer-implemented method for improved prompting of a machine-learned model. The example method can include obtaining an instructive sequence descriptive of an instructive query, an instructive response, and an instructive trace of intermediate states from the instructive query to the instructive response. The example method can include inputting, to a machine-learned model, the instructive sequence and an operative query, wherein the machine-learned model is configured to process the operative query with attention over the instructive sequence. The example method can include generating, using the machine-learned model and responsive to the operative query, an operative response.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a machine learning model using a deterministic data pipeline. One of the methods may include receiving a first request to generate a deterministic training dataset: transforming raw training examples obtained from the raw data source into pre-processed training examples; assigning a unique index to each pre-processed training example; and caching the pre-processed training examples into the cache directory specified in the received first request; receiving a second request to use the deterministic training dataset to train a machine learning model, the second request specifying a start index; and in response to receiving the second request: reading, from the cache directory, the pre-processed training examples that have indices beginning from the start index; and providing the read training examples in an order of the assigned indices for use in training the machine learning model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a machine learning model using a deterministic data pipeline. One of the methods may include receiving a first request to generate a deterministic training dataset: transforming raw training examples obtained from the raw data source into pre-processed training examples; assigning a unique index to each pre-processed training example; and caching the pre-processed training examples into the cache directory specified in the received first request; receiving a second request to use the deterministic training dataset to train a machine learning model, the second request specifying a start index; and in response to receiving the second request: reading, from the cache directory, the pre-processed training examples that have indices beginning from the start index; and providing the read training examples in an order of the assigned indices for use in training the machine learning model.

Abstract: An example method for pretraining a machine-learned model is provided. The example method includes obtaining a plurality of different combinations of configuration parameters of a pretraining objective framework. The example method includes generating, using the pretraining objective framework, a plurality of corrupted training examples from one or more training examples, wherein the plurality of corrupted training examples are respectively generated according to the plurality of different combinations. The example method includes inputting the plurality of corrupted training examples into the machine-learned model, wherein the machine-learned model is configured to generate uncorrupted subportions corresponding to corrupted subportions of the corrupted training examples. The example method includes obtaining, from the machine-learned model, a plurality of outputs respectively generated by the machine-learned model based on the plurality of corrupted training examples.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing a machine learning task on an input to generate an output. In one aspect, one of the method includes receiving input data that describes an input of a machine learning task; receiving candidate output data that describes a set of candidate classification outputs of the machine learning task for the input; generating an input sequence that includes the input and the set of candidate classification outputs; processing the input sequence using a neural network to generate a network output that specifies a respective score for each candidate classification output in the set of candidate classification outputs; and generating an output of the machine learning task for the input, comprising selecting, as the output, a selected candidate classification output from the set of candidate classification outputs using the respective scores.

Abstract: A method for processing an augmented reality (AR) workspace includes: projecting a grid comprising a cell onto a surface of the AR workspace; obtaining a first image of the AR workspace that includes the grid and a first content of the grid; extracting the first content from the first image of the AR workspace; and generating an internal representation of the AR workspace that includes the extracted first content.

Abstract: A method for converting an electronic document (ED) having a first format includes comparing an original width of the ED with a predetermined width and an original length of the ED with a predetermined length; splitting the ED into pieces in a width direction when the original width is greater than the predetermined width, and splitting the ED into pieces in a length direction when the original length is greater than the predetermined length, wherein each piece has a width and length less than or equal to the predetermined width and length; storing information representing geometric relationships of the pieces; converting the pieces from the first format into a second format; outputting the converted pieces as first output EDs; and outputting a second output ED having the second format, wherein the second output ED incorporates the first output EDs according to the information when displayed by a viewer application.

Abstract: A method for converting an electronic document (ED) having a first format includes comparing an original width of the ED with a predetermined width and an original length of the ED with a predetermined length; splitting the ED into pieces in a width direction when the original width is greater than the predetermined width, and splitting the ED into pieces in a length direction when the original length is greater than the predetermined length, wherein each piece has a width and length less than or equal to the predetermined width and length; storing information representing geometric relationships of the pieces; converting the pieces from the first format into a second format; outputting the converted pieces as first output EDs; and outputting a second output ED having the second format, wherein the second output ED incorporates the first output EDs according to the information when displayed by a viewer application.

Abstract: Methods disclosed facilitate the scheduling of print jobs. In some embodiments, the retrieval of remote files that are targets of print requests can be deferred. Print requests may be specified using JDF and/or JMF. A print queue holds entries corresponding to the remote target files and each queue entry is also associated with location information for its corresponding remote target file. When the queue entry is processed the location information associated with the queue entry is used to retrieve the remote target file for printing.

Abstract: Methods disclosed permit compositing operations to be performed on images using an associated mask even in situations where the image and the mask differ in size. In some embodiments, image and mask data may be specified as a soft mask image in a page description language such as PDF. Scaling operations may be performed on the image, mask, or on both the image and mask when they differ in size. Compositing operations may be performed on the scaled image and/or mask after they have been scaled to the same size. Composting operations in situations where the original mask and image are of the same size are not affected by scaling operations.

Abstract: Methods disclosed facilitate the scheduling of print jobs. In some embodiments, the retrieval of remote files that are targets of print requests can be deferred. Print requests may be specified using JDF and/or JMF. A print queue holds entries corresponding to the remote target files and each queue entry is also associated with location information for its corresponding remote target file. When the queue entry is processed the location information associated with the queue entry is used to retrieve the remote target file for printing.

Abstract: Methods disclosed permit compositing operations to be performed on images using an associated mask even in situations where the image and the mask differ in size. In some embodiments, image and mask data may be specified as a soft mask image in a page description language such as PDF. Scaling operations may be performed on the image, mask, or on both the image and mask when they differ in size. Compositing operations may be performed on the scaled image and/or mask after they have been scaled to the same size. Composting operations in situations where the original mask and image are of the same size are not affected by scaling operations.