Open Access. Powered by Scholars. Published by Universities.®
Articles 1561 - 1590 of 4647
Full-Text Articles in Entire DC Network
Generating Text Description For A Photo Story Using A Visual Language Model, D Shin
Generating Text Description For A Photo Story Using A Visual Language Model, D Shin
Defensive Publications Series
Automatically curated photo stories generated by a photo library application are popular. However, such stories currently do not include audio or text to enhance the story. This disclosure describes the use of a visual language model (VLM) (multimodal model) to generate text and/or audio description that provides the context of photographs included in photo stories. The VLM is provided with a system-generated prompt to generate text for a story that finds semantic consistency across a set of selected photographs. The selected photographs are processed by a vision tokenizer and the resulting tokens are fed into the VLM context window. A …
Apparatus, System, And Method For Efficiently Driving Backplanes Of Displays With Linear Control, Ajit Ninan
Apparatus, System, And Method For Efficiently Driving Backplanes Of Displays With Linear Control, Ajit Ninan
Defensive Publications Series
Circuitry may control and/or drive the brightness and/or luminance of pixels within displays. For example, an AR head-mounted display (HMD) may include and/or represent circuitry, such as a driver unit, that controls and/or drives the brightness and/or luminance of a perceptual quantizer (PQ) and/or gamma display. In one example, the circuitry may implement 8-bit counters to control the brightness and/or luminance of pixels within the PQ or gamma display. Unfortunately, such 8-bit counters may fail to enable the display to accurately reproduce perceptual brightness and/or luminance consistent with and/or under the Barten sensitivity curve and/or model. In another example, the …
Artificial Intelligence Agent For Contextual Guidance In Form Filling, Wicksell Metellus, Sindhura Balireddi, Mike Maelzer, Marim Ganaba, Erika Shiroma
Artificial Intelligence Agent For Contextual Guidance In Form Filling, Wicksell Metellus, Sindhura Balireddi, Mike Maelzer, Marim Ganaba, Erika Shiroma
Defensive Publications Series
Online forms in enterprise settings enable collection of information relevant to complex tasks such as putting together sales quotes. Form complexity can lead to errors, confusion, and/or delays as users that attempt to fill out the forms struggle to interpret the field meanings and navigate intricate processes. This disclosure describes techniques that build an understanding of forms using semantic and structural analysis, build contextual awareness based on profile data for entities completing a form and a dynamic knowledge base, and incorporate adaptive learning for error detection in forms. Techniques such as natural language processing, named entity recognition, question answering engines, …
Photo Composition Guidance Based On Scene And Subject Analysis, Kshitij Suresh Gaikar, Pushkar Mittal
Photo Composition Guidance Based On Scene And Subject Analysis, Kshitij Suresh Gaikar, Pushkar Mittal
Defensive Publications Series
This disclosure describes techniques to guide a user in the composition of photos captured by a camera device, based on automatic scene analysis and subject framing. A camera device prompts the user to pan the device across a scene surrounding the user to capture a scene photo, e.g., a 360-degree panoramic photo. Additionally, the user captures a subject photo showing an intended subject for a new photo. An artificial intelligence (AI) model is used to analyze the scene photo and the subject photo. The model outputs locations and/or poses for the subject in the scene that makes for an appealing …
Shared Interest Representation And Recommendation
Shared Interest Representation And Recommendation
Defensive Publications Series
Methods, systems, and storage media for representing and recommending shared interests are disclosed. Exemplary implementations may: determine, for each user of a group of users of a message thread associated with a messaging platform, an interest associated with the user; aggregate the interests associated with each user of the group of users; determine, based on the aggregated interests, a shared interest associated with the group; determine, based on the shared interest associated with the group, a content recommendation for the group; and provide, to at least one user of the group of users, the content recommendation.
Deep Content Understanding For Recipe Guidance Using A Large Language Model, Nagaraj Kota, Varsha Kukreja, Jatin Jindal
Deep Content Understanding For Recipe Guidance Using A Large Language Model, Nagaraj Kota, Varsha Kukreja, Jatin Jindal
Defensive Publications Series
Online recipe searches performed by users are often highly personalized; users seek specific culinary guidance, cooking methods, styles, etc. A large language model (LLM) can be used to generate recipe insights. However, an LLM can generate ungrounded results, and/or miss specific nuances of recipes along dimensions specified by the user. This disclosure describes techniques that leverage a large language model (LLM) to mine, aggregate, and summarize recipe information across multiple pages/sources and to deliver personalized, contextual, and insightful responses to user queries for recipes via a deep understanding of the pages/sources of the recipe. Recipe tips and guidance snippets are …
Physical Money Chips To Enable Digital Transactions, Ajay Prasad
Physical Money Chips To Enable Digital Transactions, Ajay Prasad
Defensive Publications Series
Users that live in regions with limited banking penetration or users with no bank accounts cannot benefit from digital payment applications that support payments, loans, and other financial transactions. This disclosure describes techniques to recharge digital payment wallets or accounts using physical money chips. The described solution bridges the gap between physical cash and digital financial systems and is suitable for users in regions with limited banking penetration. A physical money chip is provisioned with a unique identifier and has a NFC, RFID, or other code that can be read by a smartphone. Upon detection, users can perform transactions such …
Mobile Robot Navigation Based On User-Shared Movement Trajectories, Victor Carbune, Kevin Allekotte
Mobile Robot Navigation Based On User-Shared Movement Trajectories, Victor Carbune, Kevin Allekotte
Defensive Publications Series
Movement trajectories derived from timestamped location data stored locally on a user device can be a rich information source, especially for indoor spaces. However, these cannot be used on other devices such as mobile robots. Moreover, mobile robots cannot dynamically learn from user movements within the same space. This disclosure describes techniques that leverage Visual Language Models (VLMs) as a backbone to enable a device such as a mobile robot to receive historical movement trajectories of another device within the same space. The VLM backbone is utilized to transform the trajectories into lower-level actuation based on higher-level semantic interpretations related …
Earbud Design To Support Calibration Of Personalized Hrtf, Benjamin Riot, Moqian Tian, Rajeev Nongpiur
Earbud Design To Support Calibration Of Personalized Hrtf, Benjamin Riot, Moqian Tian, Rajeev Nongpiur
Defensive Publications Series
Head related transfer function (HRTF) calibration for earbuds or earphones is essential for enabling spatial audio to provide an immersive experience for users by tailoring sound to an individual’s unique ear shape and listening profile. This disclosure describes earbuds that enable the measurement of personalized HRTF along with earbud functions of listening or two-way communication. The earbuds include a loudspeaker and a microphone at different ends of an ear-book. For HRTF calibration, the microphone is placed at the ear entrance while the loudspeaker is far away, e.g., dangling below the ear, to reduce interference. Alternatively, a charging case can be …
Reducing Unwanted Facial Glow For Users Wearing Smart Glasses, Shiblee Hasan
Reducing Unwanted Facial Glow For Users Wearing Smart Glasses, Shiblee Hasan
Defensive Publications Series
Interacting with the light emitting components of the smart glasses can result in an unintended glow on the face of the glass wearer. The glow from smart glasses can make them appear conspicuous and draw unwanted attention to the user, which can be awkward in social settings and undermine the design goal of making wearable technology subtle and unobtrusive. This disclosure describes techniques to minimize facial glow caused by inward facing displays and/or other lighting elements in smart glasses by dynamically adjusting display settings in real-time based on measurements from device sensors such as inward-facing cameras, ambient light sensors, etc. …
Using Artificial Intelligence To Adjust Entity Risk Scores Based On Rule Activity, Pramod Joisha, Timothy Peacock
Using Artificial Intelligence To Adjust Entity Risk Scores Based On Rule Activity, Pramod Joisha, Timothy Peacock
Defensive Publications Series
An entity risk score can be determined based on how an entity interacts with an organization’s computing resources. The entity risk score can be adjusted based on a rule hit score (RHS), which determines how many rules are triggered by the entity’s interactions with the organization’s computing resources. A value of the RHS can be determined using a bipartite graph that includes information for a certain time interval. The RHS value, and how the RHS value change over time, can be used to alter the entity risk score.
Mixed Reality Environment Selection Using A Virtual Object, Samuel Wu
Mixed Reality Environment Selection Using A Virtual Object, Samuel Wu
Defensive Publications Series
A computer-implemented method comprising providing, within a first mixed reality environment, an environment selection user interface element; receiving, using the environment selection user interface element, an environment selection; and generating, according to the environment selection, a second mixed reality environment.
Subscriber Selection For New Service Rollout Based On Call Detail Records, Mahendra M. Periyadanam
Subscriber Selection For New Service Rollout Based On Call Detail Records, Mahendra M. Periyadanam
Defensive Publications Series
This disclosure describes techniques that make use of artificial intelligence (AI) and machine learning (ML) to intelligently select subscribers for new service rollouts based on call detail record (CDR) analysis. CDR data, obtained with user permission, is analyzed to select subscribers who frequently or infrequently use an existing service, enabling prioritizing such subscribers to rollouts of new services. The prioritized subscriber list is provided to a provisioning system to deliver configuration updates over-the-air (OTA) (or via any other mechanism, such as over the internet) to subscriber devices. Feedback from subscriber devices regarding adoption of new features is incorporated enabling refinement …
Seamless Bootstrapping For Esim Provisioning, Pavan Nuggehalli, Jayachandran C, Jean-Philippe Cormier, Amol Tuli, Sathish K
Seamless Bootstrapping For Esim Provisioning, Pavan Nuggehalli, Jayachandran C, Jean-Philippe Cormier, Amol Tuli, Sathish K
Defensive Publications Series
This disclosure describes techniques to seamlessly bootstrap embedded subscriber identity modules (eSIM) in mobile devices. In particular, the techniques enable a user to download an operating eSIM on a new or factory-reset mobile device when WiFi connectivity is unavailable. Per the techniques, in standards-agnostic bootstrapping, the platform detects the user’s choice of carrier based on the hardware SKU (stock-keeping unit) or user selection in the eSIM manager and triggers the modem to establish a bootstrap connection with the particular carrier. Alternatively, in a standards-based bootstrapping technique, the mobile device attaches to cellular (4G, 5G, LTE, etc.) network for provisioning purposes …
Mixed Reality Environment Time Selection Using A Wrist Device, Samuel Wu
Mixed Reality Environment Time Selection Using A Wrist Device, Samuel Wu
Defensive Publications Series
A computer-implemented method comprising: providing a time selection user interface element; receiving, using the time selection user interface element, a time selection; and generating, according to the time selection, a past state of a mixed reality environment, the past state depicting the mixed reality environment at a previous time corresponding to the time selection.
A Method For Remote System Management Using A "Last Will And Testament" Signaling Scheme, Hp Inc
A Method For Remote System Management Using A "Last Will And Testament" Signaling Scheme, Hp Inc
Defensive Publications Series
In the event of a catastrophic failure of a remotely management client workstation, PC, or rack mount system, a hardware component will detect the anomaly and trigger an out-of-band (OOB) network broadcast to the management plane even if power is lost. The remote management hardware relies on a backup power source to ensure enough energy for a "Last Will and Testament" message to be broadcast to a subscriber on the network.
Automated Testing Of Distributed System, Vijay Griddaluru
Automated Testing Of Distributed System, Vijay Griddaluru
Defensive Publications Series
A service definition of each service of a distributed system is modified to include metadata indicating which method of the distributed system precedes a respective method. The service definition is further modified to include additional metadata indicating one or more fields of a test request schema of a respective service and a corresponding predefined value to be used in generating a test request message to request a test of the respective service (e.g., a request generation annotation). When a request is received to test the distributed system, a test request message is generated to initiate testing of a respective service …
Times Synchronization Of Media And Transcript, Dongeek Shin
Times Synchronization Of Media And Transcript, Dongeek Shin
Defensive Publications Series
A media item and a transcript of the media item is received. A plurality of media tokens is generated based on the media item. A plurality of transcript item tokens is generated based on the transcript. A subset of the plurality of media tokens and a subset of the plurality of transcript tokens is inputted on a rolling basis into a transformer model. A sequence of probability distribution for the subset of the plurality of transcript token is obtained from the transformer model. With each submission, a transcript token is selected from the subset of the plurality of transcript tokens …
Proactive Personalized Place Suggestions In A Digital Map, Howard Yong
Proactive Personalized Place Suggestions In A Digital Map, Howard Yong
Defensive Publications Series
Digital map applications offer a location saving feature that enables users to save locations that they want to visit in the future. However, such applications do not provide proactive notifications of when a user is near a saved location, and instead, require the user to look up their lists to determine if they are near a saved location, which can cause them to miss opportunities to visit places of interest. This disclosure describes techniques to enable a digital map application to proactively provide suggestions for locations to visit that are saved by a user, by their friends, or by other …
Entity-Aware Joint Translation Of Query And Semantic Parse, Sarthak Jauhari, Saisuresh Krishnakumaran, Dinkar Vasudevan, Rahul Goel
Entity-Aware Joint Translation Of Query And Semantic Parse, Sarthak Jauhari, Saisuresh Krishnakumaran, Dinkar Vasudevan, Rahul Goel
Defensive Publications Series
Machine translation is widely utilized to translate text between language pairs. However, the accurate recognition and translation of the entities within input text remains a challenge. Current translation models often struggle to identify entities, leading to literal and incorrect translations in queries. This disclosure describes techniques, referred to as entity-aware (EaT) that enable machine translation models to recognize entities in source text, resulting in accurate and idiomatic translations. Per the techniques, joint translation of an input query and its associated semantic parse is performed. Semantically parsed information is used as supplementary data in a manner that enables the translation model …
Pre-Emptive Heating, Ventilation, And Air Conditioning System Preparation For Arriving Occupants, Anonymous
Pre-Emptive Heating, Ventilation, And Air Conditioning System Preparation For Arriving Occupants, Anonymous
Defensive Publications Series
In very cold or very hot ambient conditions, when an arriving occupant enters the vehicle and uses the Heating, Ventilation, and Air Conditioning (HVAC) system in the vehicle, there can be a significant delay until the conditioned air coming out of the vents is close to the desired temperature. To avoid this delay, a method is proposed that will pre-emptively prime the HVAC system based on the estimated arrival time of the occupant which is determined by estimating the location, trajectory, and speed of the arriving occupant. Based on the estimated time off arrival, the vehicle will then activate the …
Dynamic Pricing System For Recommerce Transaction, Madhusudan St
Dynamic Pricing System For Recommerce Transaction, Madhusudan St
Defensive Publications Series
The present disclosure related to a technique for real-time dynamic pricing recommendations. The technique includes fetching historical data and transactional information related to the plurality of products and corresponding similar products based on queries. The technique further include calculating optimal pricing recommendations based on the fetched historical data and transactional information. Subsequently, the technique include distributing price recommendations for recommerce transaction to the plurality of merchants
A Unified Multimodal Video Conference Annotator For Real-Time Llm Applications, Yu-Chung Hsiao
A Unified Multimodal Video Conference Annotator For Real-Time Llm Applications, Yu-Chung Hsiao
Defensive Publications Series
A multimodal annotator is proposed herein that provides multimodal signals including slide presentations, facial expressions, speech tones, sentiment analysis, and join/leave metadata in addition to ordinary text meeting transcripts. The multimodal annotator may enable real-time multimodal Large Language Model (LLM) applications and can be implemented as a drop-in replacement for generating text transcripts in existing LLM applications. Thus, the multimodal annotator as proposed herein can enable more advanced, multimodal artificial intelligence (AI) assistant features for that can be used across a variety of use cases, such as for video conference use cases, call center use cases, presentation video recordings, or …
Artificial Intelligence Driven Pile User Interface For Selection And Triage Of Digital Content, Ajay Prasad, Sunil Samuel, Sanika Barve
Artificial Intelligence Driven Pile User Interface For Selection And Triage Of Digital Content, Ajay Prasad, Sunil Samuel, Sanika Barve
Defensive Publications Series
The current process of selecting and managing photos via digital galleries is cumbersome and inefficient because users must manually scroll through a large number of photos, assess each photo by opening it individually, and decide relevant actions. This process can detract from the enjoyment of viewing and sharing photos. This disclosure describes techniques for streamlining the selection and organization of multiple photos into a virtual pile by dragging and dropping them into an AI-powered dynamic UI element on screen. Generative AI is applied to the photos in the pile to automate various photo-related edits and tasks. Users can expand the …
Generative Artificial Intelligence Supported Collaborative Photo Editing Based On Auxiliary Information, Ajay Prasad, Ramprasad Sedouram, Gulmohar Khan
Generative Artificial Intelligence Supported Collaborative Photo Editing Based On Auxiliary Information, Ajay Prasad, Ramprasad Sedouram, Gulmohar Khan
Defensive Publications Series
Digital photos not only provide an archive of cherished memories, but also serve as a form of self-expression when shared within a user’s social circles. Enhancing photos using digital tools allows users to express themselves in the context of the photo, to illustrate the full context and richness of the moment and environment in which a photo is captured. This is currently a tedious manual process. This disclosure describes techniques that, with user permission, automatically analyze photos and auxiliary information using a generative artificial intelligence backend to provide photo editing suggestions via a collaborative GUI to enhance photos. Collaborating seamlessly …
Automatic Speaker Prioritization During A Video Conference, Karthik Srinivas, Ramprasad Sedouram, Ajay Prasad
Automatic Speaker Prioritization During A Video Conference, Karthik Srinivas, Ramprasad Sedouram, Ajay Prasad
Defensive Publications Series
In virtual meetings, hand raise gestures are used by participants to indicate their desire to speak. When multiple participants provide hand raises, it is difficult for moderators to assign priority to different participants. This disclosure describes the use of artificial intelligence techniques, implemented with user permission, to analyze meeting content and participant actions to automatically determine priority for different speakers. The determined priority is indicated to all participants via suitable visual indicators such as color-coded user interface elements and/or adjusting a size, color, or other attribute of participant video thumbnails.
Machine Learning Based Action Recommendations To Improve Comfort, Na
Machine Learning Based Action Recommendations To Improve Comfort, Na
Defensive Publications Series
Parents of babies as well as pets need to watch their babies and pets closely and perform various actions to fulfill their needs. However, the conditions within a home are different for different babies and pets, and individuals also have different behavior patterns. Therefore, in the event of discomfort, suggested by a crying baby, it can be difficult for the parent to determine the appropriate action to take. This disclosure describes the use of machine learning based pattern matching techniques, implemented locally on a user device and with specific user permission, to learn patterns of behavior and how they relate …
Systems For Live Media Continuity, Stefan Lindmark
Systems For Live Media Continuity, Stefan Lindmark
Defensive Publications Series
Aspects and implementations of the present disclosure provide systems that keep a live media session running even if the devices providing the live media for the session become disconnected from the session. The systems keep the viewers of the live media session informed about the status of the live media session and provide an expectation that the live media will resume. In one example, a host device can provide continuity content (e.g., videos or images) to a server, the server can periodically determine if the host device has unexpectedly disconnected from the live media session, and if so, the server …
Framework Utilizing Ai/Ml To Ensure Guaranteed Reliability In 5g Standalone (Sa) Networks, Sujin Anagani, Sudeep Pandey, Solomon Ayyankulankara Kunjan, Souvik Mondal, Ashok M Bhat, Vijay Katamreddy
Framework Utilizing Ai/Ml To Ensure Guaranteed Reliability In 5g Standalone (Sa) Networks, Sujin Anagani, Sudeep Pandey, Solomon Ayyankulankara Kunjan, Souvik Mondal, Ashok M Bhat, Vijay Katamreddy
Defensive Publications Series
Service provider mobility involves large-scale deployments that serve millions of subscribers within a cluster of network functions (NFs). Even minor impairments in a deployed cluster can significantly impact a substantial number of subscribers. To address this issue, an advanced AI/ML-based failure prediction framework is proposed herein that is specifically tailored for mobility/telecom environments. This framework not only predicts impairments before they occur but also provides recommendations for actions that peer nodes can perform to prevent disruptions to existing subscribers from these potential impairments. With this framework, ML-based predictive signature is generated by NF source and shared with central entity, such …
System, Method, And Computer Program Product For Mitigating Generative Machine Learning Model Hallucination, Sunpreet Singh Arora, Minghua Xu, Anderson Clayton Alves Nascimento, Yan Zhai, Yizhen Wang
System, Method, And Computer Program Product For Mitigating Generative Machine Learning Model Hallucination, Sunpreet Singh Arora, Minghua Xu, Anderson Clayton Alves Nascimento, Yan Zhai, Yizhen Wang
Defensive Publications Series
Systems, methods, and computer program products are provided for mitigating generative model such as a large language model (LLM) hallucination. The system includes a processor configured to receive a query and input the query to a first generative model. The processor is also configured to determine candidate responses based on an output of the first generative model, generate embeddings based on the query, and retrieve data from an embedding-indexed data store based on the embeddings. The processor is further configured to input the data to a second generative model, generate a summary based on an output of the second generative …