Abstract: A surgical assembly and system are disclosed for use in performing a surgical procedure on a patient. The assembly comprises a surgical tool, the tool being arranged to receive a first signal for use in cutting or cauterizing tissue of the patient during the surgical procedure and an electrode disposed upon the tool. The assembly further comprises an electrical generator communicatively couplable with the electrode, for generating a second signal for use in generating an electrical field from the electrode proximate a site of the surgical procedure, for removing particles suspended proximate the surgical site. The assembly further comprises a controller for controlling the application of the second signal to the electrode, a sensing arrangement for sensing an activation status of the first signal, the sensing arrangement being communicatively coupled with the controller and arranged to output a sensing signal to the controller in dependence of the activation status of the first signal.

Abstract: The systems and techniques described herein relate to predicting user conversions in online advertising. Input data associated with user and advertisement features may be processed through neural networks to generate embedding representations or feature cross representations. A multi-task layer calculates probabilities associated with multiple user actions like clicks, page views, sign-ups, or purchases. Click-through and view-through conversion probabilities may be calculated to generate a score. The systems and techniques described herein perform predictions on multiple types of user actions despite data sparsity and negative transfer challenges, enhancing advertisement targeting and improving conversion metrics.

Abstract: The systems and techniques described herein relate to predicting user conversions in online advertising. Input data associated with user and advertisement features may be processed through neural networks to generate embedding representations or feature cross representations. A multi-task layer calculates probabilities associated with multiple user actions like clicks, page views, sign-ups, or purchases. Click-through and view-through conversion probabilities may be calculated to generate a score. The systems and techniques described herein perform predictions on multiple types of user actions despite data sparsity and negative transfer challenges, enhancing advertisement targeting and improving conversion metrics.

Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for predicting a conversion rate. The program and method provide for receiving, from an advertisement service, a bid to display a first advertisement at a computing device; determining, in response to receiving the bid, a set of features that relate to the first advertisement; providing the set of features to a machine learning model configured to output a predicted conversion rate for the first advertisement, the machine learning model having been trained based on multi-task learning using plural sets of features corresponding to plural second advertisements, the plural sets of features being associated with both click-through conversions and view-through conversions; and determining, based on the output of the machine learning model with respect to the set of features, the predicted conversion rate for the first advertisement.

Abstract: A surgical assembly is disclosed for use with an ultrasonic surgical device that is arranged to deliver ultrasonic vibrations to patient tissue for use in cutting or cauterizing patient tissue during a surgical procedure. The assembly comprises an ion-generating electrode arranged to receive an electrical signal for generating ions proximate a site of the surgical procedure for removing particles suspended proximate the surgical site, and a controller for controlling the application of the electrical signal to the ion-generating electrode. The assembly further comprises a sensing arrangement for sensing a presence of ultrasonic vibrations, and which is communicatively coupled with the controller and arranged to output an activation signal to the controller to cause the electrical signal to be applied to the ion-generating electrode when the sensed ultrasonic vibrations comprise a frequency within a pre-defined frequency range.

Abstract: Techniques for creating an interest graph include obtaining content items from multiple content sources and applying tailored (e.g., source-specific) preprocessing to the content items based on their respective content source. Text is extracted and salient keywords and key phrases are identified using unsupervised machine learning models. The keywords and key phrases become nodes in an interest graph, each node comprising an embedding of a keyword or key phrase in a common embedding space, with edges representing semantic similarity based on embeddings or co-engagement patterns. The graph provides an expansive, granular, and dynamic taxonomy easily adaptable to emerging interests. The interest graph overcomes limitations of conventional taxonomies that lack depth, fail to capture niche interests, and cannot adapt to reflect evolving user preferences. The described techniques construct a rich interest graph from diverse content for improved content understanding.

Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for predicting a conversion rate. The program and method provide for receiving, from an advertisement service, a bid to display a first advertisement at a computing device; determining, in response to receiving the bid, a set of features that relate to the first advertisement; providing the set of features to a machine learning model configured to output a predicted conversion rate for the first advertisement, the machine learning model having been trained based on multi-task learning using plural sets of features corresponding to plural second advertisements, the plural sets of features being associated with both click-through conversions and view-through conversions; and determining, based on the output of the machine learning model with respect to the set of features, the predicted conversion rate for the first advertisement.

Abstract: Disclosed are systems, methods, and computer-readable storage media to present content on an electronic display. In one aspect, a method includes identifying a first candidate content and a second candidate content for presentation on an electronic display, determining a first probability and a second probability that the first candidate content and the second candidate content respectively will elicit a particular type of input response, determining a first weight and a second weight based on the first probability and the second probability respectively, selecting either the first content or the second content based on the first weight and the second weight; and presenting the selected content on the electronic display.

Abstract: A candidate content item is identified for integration into a content collection. The candidate content item is associated with a first value. Using at least one machine learning model, a select value and a skip value are automatically generated for the candidate content item. The select value indicates a likelihood that the user will select the candidate content item, and the skip value indicates a likelihood that the user will bypass the candidate content item. A second value is generated for the candidate content item based on the first value, the select value, and the skip value. The candidate content item is automatically selected from a plurality of candidate content items based on the second value meeting at least one predetermined criterion. The selected candidate content item is then automatically integrated into the content collection, which is caused to be presented on a device of a user.

Abstract: Techniques for creating an interest graph include obtaining content items from multiple content sources and applying tailored (e.g., source-specific) preprocessing to the content items based on their respective content source. Text is extracted and salient keywords and key phrases are identified using unsupervised machine learning models. The keywords and key phrases become nodes in an interest graph, each node comprising an embedding of a keyword or key phrase in a common embedding space, with edges representing semantic similarity based on embeddings or co-engagement patterns. The graph provides an expansive, granular, and dynamic taxonomy easily adaptable to emerging interests. The interest graph overcomes limitations of conventional taxonomies that lack depth, fail to capture niche interests, and cannot adapt to reflect evolving user preferences. The described techniques construct a rich interest graph from diverse content for improved content understanding.

Abstract: Disclosed are systems, methods, and computer-readable storage media to present content on an electronic display. In one aspect, a method includes identifying a first candidate content and a second candidate content for presentation on an electronic display, determining a first probability and a second probability that the first candidate content and the second candidate content respectively will elicit a particular type of input response, determining a first weight and a second weight based on the first probability and the second probability respectively, selecting either the first content or the second content based on the first weight and the second weight; and presenting the selected content on the electronic display.

Abstract: Techniques for creating an interest graph include obtaining content items from multiple content sources and applying tailored (e.g., source-specific) preprocessing to the content items based on their respective content source. Text is extracted and salient keywords and key phrases are identified using unsupervised machine learning models. The keywords and key phrases become nodes in an interest graph, each node comprising an embedding of a keyword or key phrase in a common embedding space, with edges representing semantic similarity based on embeddings or co-engagement patterns. The graph provides an expansive, granular, and dynamic taxonomy easily adaptable to emerging interests. The interest graph overcomes limitations of conventional taxonomies that lack depth, fail to capture niche interests, and cannot adapt to reflect evolving user preferences. The described techniques construct a rich interest graph from diverse content for improved content understanding.

Abstract: Techniques for automated tagging of visual content are described. A pairwise model is used to encode images and videos with a vision encoder, and encode keywords and phrases from an interest graph with a text encoder. Similarity layers compare these cross-modality embeddings by calculating distance in a shared embedding space. Scores indicate the association between visual features and text. A pairwise loss function brings together matched pairs while separating non-matches during training. Scores exceeding a threshold tag content with relevant keywords and phrases. The pairwise architecture relates images and text despite limited associated text. It leverages vision-to-text understanding for accurate tagging without per-class labels. Contrastive similarity techniques associate visual patterns with textual concepts. Automated tagging organizes user-generated content by topics using this scalable cross-modality approach.

Abstract: A content request is received from a device of a user. A plurality of candidate content items is identified. Each candidate content item has a bid value. A relevancy value is automatically generated for each candidate content item. The relevancy value indicates whether the candidate content item is likely to be skipped by the user. For each candidate content item, a combined value is automatically generated by adjusting the bid value using the relevancy value generated for the candidate content item. One or more candidate content items are automatically selected based on the combined value generated for each of the one or more candidate content items. The one or more selected candidate content items are automatically integrated into at least one placeholder area among one or more pre-selected content items as part of the aggregated content. The aggregated content is presented on the device of the user.

Abstract: A chatbot system detects commercial intent during conversations with users and provides targeted advertisements. The chatbot system extracts keyword candidates from a conversation and assigns relevance scores based on the meaning of the conversation and commercial scores based on a machine learning model trained to detect commercially-related keywords. The chatbot system selects keywords based on a combination of the scores and transmits the selected keywords to advertising content servers that provide advertising content including advertisements selected using the keywords. The chatbot system displays the advertisements to the user during the conversation.

Abstract: A chatbot system for an interactive platform is disclosed. The chatbot system retrieves a conversation history of one or more conversations between a user and a chatbot from a conversation history datastore and generates one or more summarized memories using the conversation history. One or more moderated memories are generated using the summarized memories. The moderated memories are stored in a memories datastore. A user prompt is received, and a current conversation context is generated from a current conversation between the user and the chatbot. One or more memories are retrieved from the memories datastore using the current conversation context. An augmented prompt is generated using the user prompt and the one or more memories, which is communicated to a generative AI model. A response is received from the generative AI model to the augmented prompt, which is provided to the user.

Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for predicting a conversion rate. The program and method provide for receiving, from an advertisement service, a bid to display a first advertisement at a computing device; determining, in response to receiving the bid, a set of features that relate to the first advertisement; providing the set of features to a machine learning model configured to output a predicted conversion rate for the first advertisement, the machine learning model having been trained based on multi-task learning using plural sets of features corresponding to plural second advertisements, the plural sets of features being associated with both click-through conversions and view-through conversions; and determining, based on the output of the machine learning model with respect to the set of features, the predicted conversion rate for the first advertisement.

Abstract: A chatbot system for an interactive platform. The chatbot system receives a prompt from a user and determines an intent of the user using the prompt. If the chatbot system can't determine an intent, the chatbot communicates the prompt as a personality prompt to a generative AI model. If the chatbot system can determine an intent from the prompt, the chatbot system generates an API call to an additional service using the intent. The chatbot system uses values returned by the additional service to generate a hint prompt that is communicated to the generative AI model. The chatbot system receives a response from the generative AI model to the personality prompt or hint prompt and communicates the response to the user.