Walmart Apollo, LLC

Maplebear Inc.

Min Xie

Saurav Manchanda

Haixun Wang

Rahul D Sharnagat

Kannan Achan

Shishir Kumar Prasad

Zoheb Vacheri

Prithvishankar Srinivasan

Sreenivasa Prasad Sista

Shih-Ting Lin

Krishnakumar Subramanian

Jonathan Newman

Feng Quan Liu

Alessandro Magnani

Riddhima Sejpal

Suthee Chaidaroon

Lili Yuan

Gordon McCreight

Md Mansurul Alam Bhuiyan

Shuangyu Li

Qianwen Xie

Shirong Xue

Behnoush Abdollahi

Luis Manrique

Suhas Ranganath

Embodiments relate to utilizing a language model to automatically generate a novel recipe with refined content, which can be offered to a user of an online system. The online system generates a first prompt for input into a large language model (LLM), the first prompt including a plurality of task requests for generating initial content of a recipe. The online system requests the LLM to generate, based on the first prompt input into the LLM, the initial content of the recipe. The online system generates a second prompt for input into the LLM, the second prompt including the initial content of the recipe and contextual information about the recipe. The online system requests the LLM to generate, based on the second prompt input into the LLM, refined content of the recipe. The online system stores the recipe with the refined content in a database of the online system.

Using language model to generate recipe with refined content

The disclosed subject matter relates to a system and method for providing an extended search. The system generates a list of similar intent groups based on previous engagements linking queries and intents. Upon the detection of an intent in a user query, the system accesses similar intents from the similar intent group and incorporates them into the query of the product catalog in order to obtain a complete set of results. The creation of the similar intent groups uses bipartite graphs and graph embedding to identify intents close in distance.

Methods and apparatus for improving search retrieval

A computer system uses clustering and a large language model (LLM) to normalize attribute tuples for items stored in a database of an online system. The online system collects attribute tuples, each attribute tuple comprising an attribute type and an attribute value for an item. The online system initially clusters the attribute tuples into a first plurality of clusters. The online system generates prompts for input into the LLM, each prompt including a subset of attribute tuples grouped into a respective cluster of the first plurality. Based on the prompts, the LLM generates a second plurality of clusters, each cluster including one or more attribute tuples that have a common attribute type and a common attribute value. The online system maps each attribute tuple to a respective normalized attribute tuple associated with each cluster. The online system rewrites each attribute tuple in the database to a corresponding normalized attribute tuple.

Using unsupervised clustering and language model to normalize attribute tuples of items in a database

An online concierge system receives unstructured data describing items offered for purchase by various warehouses. To generate attributes for products from the unstructured data, the online concierge system extracts candidate values for attributes from the unstructured data through natural language processing. One or more users associate a subset candidate values with corresponding attributes, and the online concierge system clusters the remaining candidate values with the candidate values of the subset associated with attributes. One or more users provide input on the accuracy of the generated clusters. The candidate values are applied as labels to items by the online concierge system, which uses the labeled items as training data for an attribute extraction model to predict values for one or more attributes from unstructured data about an item.

Weakly supervised extraction of attributes from unstructured data to generate training data for machine learning models

This application relates to apparatus and methods for providing recommended items based on a search request. In some examples, a computing device determines a plurality of items for recommendation based on an item type in the search request from a user. The computing device then determines a model variant from a plurality of variants for each item based on a requested variant in the search request. The computing device generates item recommendations by modifying an item title for each of the plurality of items to indicate the corresponding model variant. The computing device then presents the item recommendations for the user's perusal.

System, method, and non-transitory computer readable medium for generating recommendations

An online concierge system trains a classification model as a domain adversarial neural network from training data labeled with source classes from a source domain that do not overlap with target classes from a target domain output by the classification model. The online concierge system maps one or more source classes to a target class. The classification model extracts features from an image, classifies whether an image is from the source domain or the target domain, and predicts a target class for an image from the extracted features. The classification model includes a gradient reversal layer between feature extraction layers and the domain classifier that is used during training, so the feature extraction layers extract domain invariant features from an image.

Training a classification model using labeled training data that does not overlap with target classifications for the classification model

A server receives a plurality of product data entries from a plurality of retailer computing systems. Each product data entry includes a product identifier uniquely identifying a corresponding physical product and descriptive data of the corresponding physical product. A subset of the plurality of product data entries having a same product identifier is determined. An embedding vector representative of a product data entry in the subset is pairwise compared with each of respective embedding vectors representative of other product data entries in the subset other than the product data entry to compute respective vector similarity metrics. A pooled semantic similarity metric for the product data entry based on the computed respective vector similarity metrics. It is determined whether the product data entry is an outlier in the subset based on the pooled semantic similarity metric. A notification is transmitted to a client device of a user based on the determination.

Verifying association between physical products and identifiers of physical products

This application relates to systems and methods for automatically generating experiments based on experiment requests routed to micro-services (model sub-components) using a prefix-based routing mechanism. In some examples, experiment requests may parsed to determine lower layer services (e.g., components) whose properties need to be changed for a model iteration. Prefixes in requests may be used to route the experiment requests and portions thereof to appropriate services or layers for configuration at the micro-service level. Routing tables at each higher layer may be utilized to determine the correct sub-layers to redirect a request and/or portion thereof. At micro-service level, each micro-service may store and use a configuration table to match a received parameter in a request with a property and its corresponding value for the experiment.

Systems and methods for testing models

A system for retrieving products in response to a customer query includes a computing device configured to obtain query information characterizing a query initiated by the customer on an ecommerce marketplace and to determine embedding-based search results comprising a first list of items. The computing device is also configured to obtain legacy search results comprising a second list of items and to blend the embedding-based search results with the legacy search results to obtain blended search results. The computing device is also configured to send the blended search results to the customer.

Methods and apparatuses for providing search results using embedding-based retrieval

Systems and methods including one or more processing modules and one or more non-transitory storage modules storing computing instructions configured to run on the one or more processing modules and perform acts of training one or more first models to recommend a first item after a user has had an interaction on the web site of the online retailer with a second item, determining static features common to both the first item and the second item, training a second model to determine whether to coordinate a display of any new item as one of one or more recommended items with any of a plurality of items, and coordinating the display of the new item as one of the one or more recommended items when the one or more of the plurality of items are displayed on the website of the online retailer based on the static features of the new item.

Growing community of inventors

Foster City, CA, United States of America

Min Xie