Ancestry.com Operations Inc.

The present disclosure is directed toward systems, methods, and non-transitory computer-readable media for generating bounding boxes for an image in accordance with one or more embodiments. The disclosed systems can generate a set of hidden states for an image of a document. The disclosed systems can generate a set of bounding boxes and a set of tokens from the hidden states utilizing a first head component and a second head component of a record generation model. The disclosed systems can aggregate a plurality of bounding boxes into a bounding box group. Additionally, the disclosed systems can provide the image depicting the bounding box group for display within a graphical user interface of a client device. Additionally or alternatively, the disclosed systems can train the record generation model, in part, by determining that a loss associated with masking one or more training bounding boxes satisfies a threshold loss.

The present disclosure is directed toward systems, methods, and non-transitory computer-readable media for a retrieval augmented thought approach to generating responses using neural networks. For example, the disclosed systems can receive a prompt instruction a large language model to generate a response. The disclosed systems can identify a retrieval-augmented-thought (RAT) store item corresponding to the prompt from among RAT-store items stored in a repository accessible by the large language model. In some cases, the disclosed systems can retrieve the RAT-store item and can generate a response using the large language model to execute processes defined by the RAT-store item. Further, in some cases, the disclosed systems utilize a dynamic graph-based agentic framework that incorporates one or more functional adapters and a RAT replanner to execute a sequence of processes indicated by a RAT-store item in a sequential manner.

Disclosed herein relates to a method that analyzes the sentiment of user feedback for a genealogical system and identifies key phrases that may relate to novel themes in the user feedback. Sentiment analysis and novel theme prediction systems, methods, and computer-program products are described. Sentiment analysis of user feedback may include dividing user-generated unstructured text files into sections. The method classifies each section to an aspect of the genealogical system from a predetermined list of aspects monitored by the genealogical system. The method inputs the text belonging to the classified section to a supervised machine learning model and determines a sentiment associated with the classified section. In other embodiments, a method generates embedding vectors representing survey responses from users of a genealogical system. The method extracts a subset of survey responses having embedding vectors grouped into one cluster. The method extracts key phrases that may indicate a novel theme.

Described herein are systems, methods, and other techniques for extracting one or more keyphrases from an input text. The input text may include a plurality of words. A plurality of token-level attention matrices may be generated using a transformer-based machine learning model. The plurality of token-level attention matrices may be converted into a plurality of word-level attention matrices. A set of candidate phrases may be identified from the plurality of words based on the plurality of word-level attention matrices. The one or more keyphrases may be selected from the set of candidate phrases.

Described herein are systems, methods, and other techniques for segmenting an input text. A set of tokens are extracted from the input text. Token representations are computed for the set of tokens. The token representations are provided to a machine learning model that generates a set of label predictions corresponding to the set of tokens. The machine learning model was previously trained to generate label predictions in response to being provided input token representations. Each of the set of label predictions indicates a position of a particular token of the set of tokens with respect to a particular segment. One or more segments within the input text are determined based on the set of label predictions.

Disclosed herein are methods, systems, and non-transitory computer readable mediums for generating a shareable genealogical summary for a target individual. An example method includes receiving a request from a user to generate a shareable genealogical summary about a target user. The method generates the shareable genealogical summary comprising a genealogical history of the target user. The method provides genealogical information for the target user to a machine-learning language model. The genealogical information includes a family tree. The method receives a response generated by executing the machine-learning language model from a model serving system. The method provides the shareable genealogical summary for display to the user.

Systems and methods for handwriting recognition using language modeling facilitate improved results by using a trained language model to improve results from a handwriting recognition machine learning model. The language model may be a character-based language model trained on a dataset pertinent to field values on which the handwriting recognition model is to be used. A loss prediction module may be trained with the handwriting recognition model and/or the language model and used to determine whether a prediction from the handwriting recognition model should be refined by passing the prediction through the trained language model.

Described herein are systems, methods, and other techniques for segmenting an input text. A set of tokens are extracted from the input text. Token representations are computed for the set of tokens. The token representations are provided to a machine learning model that generates a set of label predictions corresponding to the set of tokens. The machine learning model was previously trained to generate label predictions in response to being provided input token representations. Each of the set of label predictions indicates a position of a particular token of the set of tokens with respect to a particular segment. One or more segments within the input text are determined based on the set of label predictions.

Disclosed herein relates to a method that analyzes the sentiment of user feedback for a genealogical system and identifies key phrases that may relate to novel themes in the user feedback. Sentiment analysis and novel theme prediction systems, methods, and computer-program products are described. Sentiment analysis of user feedback may include dividing user-generated unstructured text files into sections. The method classifies each section to an aspect of the genealogical system from a predetermined list of aspects monitored by the genealogical system. The method inputs the text belonging to the classified section to a supervised machine learning model and determines a sentiment associated with the classified section. In other embodiments, a method generates embedding vectors representing survey responses from users of a genealogical system. The method extracts a subset of survey responses having embedding vectors grouped into one cluster. The method extracts key phrases that may indicate a novel theme.

Described herein are systems, methods, and other techniques for segmenting an input text. A set of tokens are extracted from the input text. Token representations are computed for the set of tokens. The token representations are provided to a machine learning model that generates a set of label predictions corresponding to the set of tokens. The machine learning model was previously trained to generate label predictions in response to being provided input token representations. Each of the set of label predictions indicates a position of a particular token of the set of tokens with respect to a particular segment. One or more segments within the input text are determined based on the set of label predictions.