Alerta

SYSTEM AND METHODS OF OPTIMIZING INFERENCE FOR FOUNDATION MODELS

Absstract of: WO2026123313A1

A computer-implemented method is disclosed. The method includes: receiving a request to process using a pre-trained machine learning model, the request including an input prompt, a first budget, and a task identifier of a first task; pruning the machine learning model, without storing a separate pruned model in memory, wherein the pruning comprises determining, for each layer of the machine learning model, a layer gate value indicating whether to use the layer for inference based on the first budget and the first task; and utilizing the pruned machine learning model in accordance with the layer gate values to generate an output corresponding to the request.

CLOSED-LOOP OPTIMIZATION OF GENERAL REACTION CONDITIONS FOR HETEROARYL SUZUKI-MIYUARA COUPLING

Absstract of: US20260170410A1

0000 Disclosed are systems and methods for rapidly generating general reaction conditions using a closed-loop workflow leveraging matrix down-selection, machine learning, and robotic experimentation.

MULTISPECTRAL ANALYSIS USING A SMARTPHONE CAMERA FOR MEASURING CONCENTRATIONS OF LIGHT-EMITTING COMPOUNDS

Absstract of: US20260170715A1

In some embodiments, a computer-implemented method of measuring light-emitting compounds using a smartphone is provided. The smartphone determines a transformation matrix using one or more options specified via a configuration user interface. The smartphone transforms a low-color space image that depicts at least a subject into a multispectral data cube using the transformation matrix, and determines a measurement of a light-emitting compound associated with the subject using the multispectral data cube. In some embodiments, a computer-implemented method of measuring light-emitting compounds using a smartphone is provided. The smartphone transforms the low-color space image that depicts at least a subject into a multispectral data cube using a transformation matrix. The smartphone provides values from the multispectral data cube to an ensemble of two or more machine learning models, and determines a measurement of a light-emitting compound associated with the subject based on outputs of two or more machine learning models.

CHARACTERISTIC-BASED PREDICTIVE OPERATIONAL ASSIGNMENT

Absstract of: US20260169468A1

Techniques are provided for determining elements of a routing. A set of inputs is obtained, where respective inputs are associated with sets of one or more characteristics. Values for the characteristics are associated with a set of labels defining operational routing attributes and are used to train a machine learning model. Inference data including characteristic values for inputs is analyzed using the machine learning model to produce an inference result identifying predicted labels. The predicted labels are used to execute at least a portion of a routing operation including assignments of work centers, execution sequences, or resources. Updated inference data may result in different predicted labels and different routing assignments. Using characteristic values enables improved inference accuracy and allows a greater portion of available data to be used for training.

CONTEXTUALLY RECOMMENDING LEVELS OF DRIVING AUTOMATION THROUGH MULTI-VEHICLE COLLABORATION

Absstract of: US20260170954A1

A method includes: receiving contextual data from an autonomous vehicle, wherein the contextual data includes data obtained from the autonomous vehicle and data obtained from other vehicles; in response to receiving the contextual data from the autonomous vehicle, determining a level of driving automation that is predicted to provide a highest level of efficiency of the autonomous vehicle based on the contextual data and using a machine learning model; and transmitting, to the autonomous vehicle, a recommendation including the determined level of driving automation.

COMPUTER-IMPLEMENTED METHOD OF AUTOMATED TRAVEL ITINERARY PLANNING

Absstract of: US20260170074A1

0000 The present invention relates to an automated travel planning data processing system and method that leverages advanced artificial intelligence (Al) and machine learning (ML) algorithms to generate personalized travel itineraries in real-time. The system comprises a central server with one or more processors, memory, and a machine learning module, as well as a travel database that stores aggregated data from multiple sources. Users interact with the system through a natural language processing-based user interface, which receives inputs comprising travel dates, destinations, and preferences. An Al-powered itinerary generation engine processes user inputs and aggregated data to create personalized travel plans, utilizing a multithreading module for simultaneous data retrieval and processing. The machine learning module continuously optimizes the itinerary generation process by analyzing user preferences and travel data patterns. The invention also provides a method for automated travel itinerary planning using the data processing system.

CLUSTER AND LANGUAGE-BASED FORECASTING FOR NUMERIC TIME SERIES

Absstract of: US20260170292A1

In an example embodiment, a large language model (LLM) is utilized to generate a semantic vector for each given time series. These semantic vectors represent additional information generated based on descriptions of the type of the time series (e.g., a description of the material, whose demand over time comprises the time series). The semantic vectors can then be used to stabilize the assignment of clusters in a cluster-based machine learning model, especially for short time series, to improve reliability of predictions.

SOFTWARE SOLUTION AND PLATFORM THAT INTEGRATES AND COMBINES MULTIPLE INTERCHANGEABLE SOFTWARE COMPONENTS USING CUSTOM DATA PIPES

Absstract of: US20260169703A1

Disclosed herein is a software solution and platform that integrates and combines multiple interchangeable software components using custom data pipes. The software solution and platform, as well as the multiple interchangeable software components, can include machine learning (ML), artificial intelligence (AI), and generative AI. The software solution and platform is a processor executable code or software that is necessarily rooted in process operations by, and in processing hardware of, computing equipment. For ease of explanation, the software solution and platform is described herein with respect to an integration engine.

SYSTEM AND METHOD FOR PREDICTING A PHYSICAL FIELD

Absstract of: US20260170331A1

0000 The method can include determining a physical problem; determining an initial solution for the physical problem description; determining a final solution of the physical problem description, using the initial solution as a seed; optionally training a machine learning model based on the final solution. The method functions to quickly determine a high-quality solution (e.g., accurate and/or precise solution) to a physical problem by leveraging a predictive machine learning model for determining initial approximate solutions to the physical problem.

POLICY-CENTRIC DATA COLLECTION FOR DISTRIBUTED MACHINE-LEARNING

Absstract of: US20260170400A1

A computer system can: identify a training target associated with training a machine-learned model; communicate, to a remote computing device, a data capture policy that is implementable by the remote computing device to cause the remote computing device to capture data responsive to the training target; obtain a component training dataset from the remote computing device, the component training dataset comprising data captured according to the data capture policy; train the machine-learned model using an aggregate training dataset comprising the component training dataset aggregated with a plurality of additional component training datasets from a plurality of additional remote computing devices to generate an update for the machine-learned model; and communicate an update for the machine-learned model to the remote computing device for updating a local instance of the machine-learned model at the remote computing device.

SYSTEM AND METHOD OF GENERATING DIGITAL ITEM RECOMMENDATIONS

Absstract of: US20260172613A1

0000 A method and a server for generating item recommendations for users of a digital recommendation platform are provided. The method includes training a machine-learning algorithm (MLA) to identify a next digital item from to be provided to a given user. The training includes: identifying a set of least interacted digital items; determining, for a given least interacted digital item, at least one prior user having interacted therewith; generating, based on respective pluralities of user features of the at least one prior user, a respective auxiliary plurality of item features; augmenting the respective plurality of item features of the given least interacted item with the respective auxiliary plurality of item features, thereby generating a respective augmented plurality of item features for the given least interacted digital item; and training the MLA based on the respective augmented pluralities of item features of the set of least interacted digital items.

AUTOMATED EXPERIMENTATION AND EVALUATION OF MACHINE LEARNING SOLUTIONS

Absstract of: WO2026128383A1

A document is determined to be relevant to a machine learning problem. An autonomous research agent is utilized using a solution to the machine learning problem to conduct an experiment related to the machine learning problem based on information included in the document. The conducted experiment is evaluated. The solution to the machine learning problem is updated based on the evaluation of the conducted experiment.

OUTCOME-MAPPED DECISION MODEL FOR NETWORK ROUTING

Absstract of: US20260172341A1

A method and related systems may generate and use a machine learning model that determines multiple outcome values for multiple message routes. Some embodiments may then compare the multiple outcome values to select a target route for the message. Moreover, some embodiments may facilitate routing through one or more networks by mapping aggregated performance metrics through controlled nodes.

UNSTRUCTURED ASSET CONVERGENCE, MONITORING, AND FORECASTING

Absstract of: WO2026128626A1

Disclosed are systems, apparatuses, processes, and computer-readable media for unstructured asset convergence, monitoring, and forecasting. For example, a disclosed method includes receiving a collection of unstructured documents; generating structured data based on the collection of unstructured documents; identifying a property identifier of a first asset within the structured data, the property identifier being associated with a single physical structure; in response to receiving an unstructured notification, identifying an entity associated with the unstructured notification and a second asset corresponding to the entity; determining, by a machine learning model, that the first asset will be affected based on the unstructured notification based on at least one of the entity, a proximity of the first asset to the second asset; and providing a notification regarding the first asset and potential changes to the first asset based on the entity or the second asset.

Income Inference And Validation Using Layered Machine Learning Models

Absstract of: US20260170554A1

0000 The present disclosure relates generally to an income inference and validation system. For example, the system can receive application data and a stated income. The system can further identify borrower attributes based on the application data and obtain table features. The table features can include subsets of the borrower attributes and corresponding feature values. The system can ingest the table features into a first machine learning (ML) model to generate a predicted income for the application. Additionally, the system can generate, by a second ML model in which the predicted income and the table features are ingested, an error estimate for the predicted income. The system can further ingest the predicted income, the error estimate, and the table features into a third ML model. The system can generate, with the third ML model an income score indicating a likelihood that the stated income is fraudulent.

STORING A MACHINE LEARNING MODEL USING DIFFERENTIAL STORAGE OF TENSOR DATA

Absstract of: US20260170396A1

The technologies described herein are generally directed toward storing machine learning model data using differential storage of tensor data. For instance, a system can enable performance of operations including storing, as data chunks in storage, a first tensor version including first model parameters of a machine learning model and, after the first tensor version was stored, the first model parameters have been changed to second model parameters of the machine learning model, resulting in a second tensor version that includes the second model parameters. The method may further include, based on a command to checkpoint the machine learning model, determining a changed data chunk of the data chunks corresponding to a change from the first tensor version to the second tensor version. Further, the method may include storing the changed data chunk as a checkpoint of the machine learning model.

MACHINE LEARNING MODEL EVALUATION METHOD, DATA PROCESSING METHOD, AND RELATED DEVICE

Absstract of: US20260170420A1

0000 A machine learning model evaluation method, a data processing method, and a related device are disclosed. The method can be applied to the autonomous driving field of artificial intelligence. The method includes: processing a plurality of pieces of segmented data in an evaluation sample using a machine learning model, to generate a plurality of prediction labels, and determining a parameter value of at least one evaluation indicator. The evaluation sample includes description data of a traffic scene within a first duration, the segmented data includes description data of the traffic scene within a first sub-duration of the first duration. The at least one evaluation indicator indicates stability and/or accuracy corresponding to the plurality of prediction labels, and the accuracy is obtained based on ground-truths corresponding to the plurality of pieces of segmented data.

INTERACTION PROFILING OF MACHINE LEARNING MODEL LAYERS

Absstract of: US20260170404A1

Techniques for detecting security threats/vulnerabilities resulting from changes to a machine learning (ML) model during operation are disclosed. An initial interaction graph that maps each component of an ML model to system processes of a host operating system (OS) that the component interacts with may be generated. Changes in the ML model may be monitored and in response to detecting a change in the ML model that meets a magnitude criteria, an updated interaction graph may be generated. The updated interaction graph may map each component of the changed ML model to the system processes of the host OS that the component interacts with. The initial interaction graph and the updated interaction graph are compared to determine an interaction graph delta, and one or more mitigating actions are taken based on the interaction graph delta.

Taxonomic Negative Sampling-Based Machine Learning

Absstract of: US20260170394A1

Techniques for generating negative samples using a taxonomy of entities and relations and for using the negative samples for machine learning are disclosed herein. Taxonomic negative samples are generated by selecting entities and negated relations for the entities using a taxonomy of entity types and subject, object relations. The system defines taxonomic negative samples for same-sentence relations, cross-sentence relations, and/or header context relations. The system sieves taxonomic negative samples are sieved to eliminate some samples, such as false negative samples. The sieved taxonomic negative samples are included with positive samples in training data used to train and/or fine-tune a prediction engine or other machine learning model.

MULTIMODAL MACHINE-LEARNING MODEL FOR PREDICTING ITEM USAGE

Absstract of: US20260170405A1

0000 An online system trains a multimodal machine-learning model to predict a rate of using an item that can be ordered at the online system by a user. The machine-learning model is trained by using a plurality of training examples, where each training example includes training images associated with a respective training user that are related to a respective item from the collection of items, and data related to conversion of the respective item by the respective training user. Upon receiving images of user's physical spaces that store items, the online system applies the trained machine-learning model to the images to output a rate of using a specific item by the user. Based on the predicted rate, the online system generates a user interface signal causing a device associated with the user to display a user interface with a user interface element for use by the user to restock the item.

METHOD FOR MEASURING SPATIAL ANGLE OF DRAG PIPE OF DREDGER

Absstract of: WO2026124029A1

A method for measuring a spatial angle of a drag pipe of a dredger, which relates to the technical field of ship engineering. The method comprises: collecting motion data of a hull and motion data of a drag pipe in real time, and using median filtering to remove high-frequency noise; extracting features, and constructing a comprehensive feature vector; using a fuzzy inference system to decouple nonlinear influences between the shaking of the hull and the attitude of the drag pipe; correcting an attitude estimation result on the basis of real-time error feedback; dynamically optimizing the estimation result by means of Bayesian estimation and particle filtering algorithms; and correcting a system deviation. Therefore, the measurement precision is improved. A system further evaluates the overall measurement accuracy on the basis of a machine learning model, and automatically adjusts measurement parameters and processes, so as to improve the adaptability and robustness of the system; and for an inaccurate measurement result, the system can analyze the cause and re-perform spatial angle measurement, thereby ensuring long-term stable operation.

ANOMALY DETECTION BASED ON ENSEMBLE MACHINE LEARNING MODEL

Absstract of: US20260170419A1

0000 A security platform employs a variety techniques and mechanisms to detect security related anomalies and threats in a computer network environment. The security platform is “big data” driven and employs machine learning to perform security analytics. The security platform performs user/entity behavioral analytics (UEBA) to detect the security related anomalies and threats, regardless of whether such anomalies/threats were previously known. The security platform can include both real-time and batch paths/modes for detecting anomalies and threats. By visually presenting analytical results scored with risk ratings and supporting evidence, the security platform enables network security administrators to respond to a detected anomaly or threat, and to take action promptly.

METHOD FOR GENERATING TRAINED PREDICTION MODEL THAT PREDICTS ENERGY EFFICIENCY OF MELTING FURNACE, METHOD FOR PREDICTING ENERGY EFFICIENCY OF MELTING FURNACE, AND COMPUTER PROGRAM

Absstract of: US20260168731A1

0000 A method of generating a trained model includes: a step of acquiring a process state parameter for every single charge (S110); a step of performing preprocessing by applying machine learning to a data set of one or more process state parameters acquired through m charges (where m is an integer of 2 or greater) (S130); a step of generating a learning data set (S140); and a step of generating a trained model (S150). The learning data set is generated based on n-dimensional features (where n is an integer of 1 or greater) that have been extracted through the preprocessing, and at least contains one or more process target parameters representing process fundamental information that is set for every single charge.

METHOD AND SYSTEM TO PROVIDE REAL TIME INTERIOR ANALYTICS USING MACHINE LEARNING AND COMPUTER VISION

Absstract of: US20260170870A1

0000 A system may be configured to generate real-time analytics using machine learning and computer vision. In some aspects, the system may receive a first video frame from a first video capture device positioned to capture activity at a location within a monitored area, receive a second video frame from a second video capture device positioned to capture interaction activity of customers, determine a first inference based on the first video frame, determine a second inference based on the second video frame, and associate the first inference and the second inference based on determining that the first inference and the second inference correspond to a common time period and common location.

MACHINE LEARNING MODELS FOR GAPS IN CARE AND MEDICATION ACTIONS

Nº publicación: US20260171247A1 18/06/2026

Applicant:

CLOVER HEALTH INVEST CORP [US]
Clover Health Investments Corp.

Absstract of: US20260171247A1

0000 The present disclosure describes methods and systems for machine learning models utilized for identifying issues with medications and gaps in care. The present disclosure also describes methods and systems for machine learning models utilized to provide recommended actions for addressing the issues with medication and gaps in care. These methods and systems utilize machine learning models that are trained to identify issues with medications and gaps in care, as well as provide recommended actions for those medications issues and gaps in care. The models are trained with data from disparate sources that are aggregated and formatted to be utilized in these models.