Abstract: A dataset is accessed that is to be classified into topics. A subset of the dataset is selected and used to generate themes using a language model. Each item in the subset is classified and labeling into the set of themes using the language model. A classifier model is trained using the classified and labeled subset and the generated themes. The trained classifier model is used to classify the dataset into the set of themes.

Abstract: A set of data is accessed that is to be classified into topics. The topics are generated across multiple data partitions of the set of data using concurrent calls to a first language model. The generated topics are consolidated to generate a set of topics. Each item in the set of data are classified into the set of topics using parallel calls to a second language model. An output is generated identifying which items of the set of data are classified into which topics of the set of topics.

Abstract: Systems and methods are directed to building annotated models based on eyes-off data. Specifically, a synthetic data generation model is trained and used to further train a target model. The synthetic data generation model is trained within an eyes-off environment using an anonymity technique on confidential data. The synthetic data generation model is then used to create synthetic data that closely represents the confidential data but without any specific details that can be linked back to the confidential data. The synthetic data is then annotated and used to train the target model within an eyes-on environment. Subsequently, the target model is deployed back within the eyes-off environment to classify the confidential data.

Abstract: A computer-implemented method for obtaining actionable application feedback includes executing an application on remote computing systems and, during execution of the application on each computing system: surfacing a UI including a survey UI element on the display of the computing system; surfacing a first survey prompt via the survey UI element; receiving user input including a response to the first survey prompt, inputting the user input to an LLM; generating, via the LLM, a second survey prompt based on the provided user input; surfacing the second survey prompt via the survey UI element; receiving user input including a verbatim response to the second survey prompt; and generating, via the LLM, topic tag(s) corresponding to the verbatim response. The method includes aggregating the verbatim responses received via the computing systems according to the corresponding topic tag(s), as well as generating, via the LLM, application insights based on the aggregated verbatim responses.

Abstract: A method and system for generating synthetic privacy preserving training data for training a language classifier machine-learning (ML) model includes receiving a request to generate the synthetic privacy-preserving training data for the language classifier ML model, retrieving labeled training data associated with training the language classifier ML model, providing the labeled training data, one or more privacy parameters, and a domain type associated with the labeled training data to a synthetic data generation ML model, the synthetic data generation ML model being configured to generate synthetic training data in a privacy-persevering manner, receiving synthetic privacy-preserving training data as an output from the synthetic data generation ML model, and providing the synthetic privacy preserving training data to the language classifier ML model for training the language classifier ML model in classifying text.

Abstract: Systems and methods are directed to building annotated models based on eyes-off data. Specifically, a synthetic data generation model is trained and used to further train a target model. The synthetic data generation model is trained within an eyes-off environment using an anonymity technique on confidential data. The synthetic data generation model is then used to create synthetic data that closely represents the confidential data but without any specific details that can be linked back to the confidential data. The synthetic data is then annotated and used to train the target model within an eyes-on environment. Subsequently, the target model is deployed back within the eyes-off environment to classify the confidential data.

Abstract: Systems and methods are directed to building annotated models based on eyes-off data. Specifically, a synthetic data generation model is trained and used to further train a target model. The synthetic data generation model is trained within an eyes-off environment using an anonymity technique on confidential data. The synthetic data generation model is then used to create synthetic data that closely represents the confidential data but without any specific details that can be linked back to the confidential data. The synthetic data is then annotated and used to train the target model within an eyes-on environment. Subsequently, the target model is deployed back within the eyes-off environment to classify the confidential data.

Abstract: A method and system for generating synthetic privacy preserving training data for training a language classifier machine-learning (ML) model includes receiving a request to generate the synthetic privacy-preserving training data for the language classifier ML model, retrieving labeled training data associated with training the language classifier ML model, providing the labeled training data, one or more privacy parameters, and a domain type associated with the labeled training data to a synthetic data generation ML model, the synthetic data generation ML model being configured to generate synthetic training data in a privacy-persevering manner, receiving synthetic privacy-preserving training data as an output from the synthetic data generation ML model, and providing the synthetic privacy preserving training data to the language classifier ML model for training the language classifier ML model in classifying text.