Intuit, Inc.

Maplebear Inc.

Xiao Xiao

Taesik Na

Haixun Wang

Tejaswi Tenneti

Dominic Miguel Rossi

Madelaine Daianu

Saisri Padmaja Jonnalagedda

Yao H Morin

Sricharan Kallur Palli Kumar

Xuan Zhang

Peter Ouyang

Homa Foroughi

Jonathan Lunt

Li Tan

Vinesh Reddy Gudla

Ling Feng Wei

Andre Felipe Luis

Itai Jeczmien

Jian Li

Allegra Latimer

Ayantha Randika Ponnamperuma Arachchige

Nilanjan Ray

Charles Showley

Ruhan Zhang

Xiaochen Wang

An online system updates the labels on negative examples to account for the possibility that the example is a false negative. The system generates a set of initial training examples that each include a query input by the user and item data for an item presented as a result to the user's query. Each training example also includes an initial label, which represents whether the user interacted with the item presented as a search result. The online system updates the initial label for a negative training example by identifying a set of bridge queries and computing a similarity score between the query for the training example and the bridge queries. The online system computes an updated label for the negative example based on the similarity scores and updates the training example with the updated label.

False negative prediction for training a machine-learning model

An online system leverages stored interactions with items made by users after the online system received queries to determine display of items satisfying the query. For example, the online system trains a model to predict a likelihood of a user performing an interaction with an item displayed after a query was received. As different items receive different amounts of interaction from users, limited historical interaction with certain items may limit accuracy of the model. The online system generates embeddings for previously received queries and uses measures of similarity between embeddings for queries to generate clusters of queries. Previous interactions with queries in a cluster are combined, with the combined data being used for determining display of items in response to a query.

Clustering data describing interactions performed after receipt of a query based on similarity between embeddings for different queries

A system may generate a prompt based in part on a search query from a customer client device. The prompt instructs a machine learned model to provide item predictions. And the model was trained by: converting structured data describing items of an online catalog to annotated text data (unstructured data), generating training examples based in part on the annotated text data, and training the model using the training examples. The system may receive item predictions generated by the prompt being applied to the machine learned model, the item predictions may have corresponding item identifiers. The item predictions are processed to identify a recommended item from the item predictions. The processing includes determining item information for the recommended item using an item identifier associated with the recommended item. The item information is provided to the customer client device.

Machine learned models for search and recommendations

An online system displays search results in response to a query by receiving a query from a customer. An online system accesses a set of candidate items and computes a relevance score and personalization score for each item. The online system computes the relevance score based on query data and item data and may normalize the relevance score. The online system computes the personalization score based on item data, such as an item embedding, and user data, such as a user embedding. The online system computes a query specificity score and adjusts the personalization score with the query specificity score such that generic queries have high personalization scores and specific queries have low personalization scores. The online system combines the relevance and personalization scores for each candidate item into a ranking score and displays the candidate items to the customer based on their ranking scores.

Providing and displaying search results in response to a query

A processor may receive an image and identify a plurality of characters in the image using a machine learning (ML) model. The processor may generate at least one word-level bounding box indicating one or more words including at least a subset of the plurality of characters and/or may generate at least one field-level bounding box indicating at least one field including at least a subset of the one or more words. The processor may overlay the at least one word-level bounding box and the at least one field-level bounding box on the image to form a masked image including a plurality of optically-recognized characters and one or more predicted fields for at least a subset of the plurality of optically-recognized characters.

End to end trainable document extraction

A computing system receives, from a client device, an image of a content item uploaded by a user of the client devices. The computing system divides the image into one or more overlapping patches. The computing system identifies, via a first machine learning model, one or more distortions present in the image based on the image and the one or more overlapping patches. The computing system determines that the image meets a threshold level of quality. Responsive to the determining, the computing system corrects, via a second machine learning model, the one or more distortions present in the image based on the image and the one or more overlapping patches. Each patch of the one or more overlapping patches are corrected. The computing system reconstructs the image of the content item based on the one or more corrected overlapping patches.

Machine learning pipeline for document image quality detection and correction

A method includes executing a Optical Character Recognition (OCR) preprocessor on training images to obtain OCR preprocessor output, executing an OCR engine on the OCR preprocessor output to obtain OCR engine output, and executing an approximator on the OCR preprocessor output to obtain approximator output. The method further includes iteratively adjusting the approximator to simulate the OCR engine using the OCR engine output and the approximator output, and generating OCR preprocessor losses using the approximator output and target labels. The method further includes iteratively adjusting the OCR preprocessor using the OCR preprocessor losses to obtain a customized OCR preprocessor.

Preprocessor training for optical character recognition

Certain aspects of the present disclosure provide techniques for determining query intent and complexity based on text input. One example method generally includes receiving, from a user device, a text query and preprocessing the text query to generate a query vector. The method further includes providing the query vector to an intent model configured to output a user intent of the text query and providing the query vector to a complexity model configured to output a complexity of the text query. The method further includes receiving the user intent of the text query from the intent model and receiving the complexity of the text query from the complexity model. The method further includes determining, based on the user intent and the complexity of the query, a routing target for the text query and routing the text query to the routing target.

Determining query intent and complexity using machine learning

Certain aspects of the present disclosure provide techniques for providing a routing system to a user of a product. An example technique includes receiving from a user of a product a query and a personal ID. Based on the personal ID of the user, the user's profile is retrieved which comprises user attribute data, a clickstream history of the user, and a product SKU of the product. Based on the query and the user profile, processed user data is generated. Additionally, agent profile data for each available agent is retrieved, and based on the user attribute data, the processed user data, and the agent profile data of each agent, a predicted quality score is generated for each agent. The agent with the highest predicted quality score is determined, and the user is routed to the agent with the highest predicted quality score.

Routing system to connect a user with a qualified agent

Growing community of inventors

San Diego, CA, United States of America

Xiao Xiao