Alerta

HANDLING DATABASE OBJECT NAMING IN NL2SQL WITH LARGE LANGUAGE MODELS

Resumen de: US20260289393A1

Techniques are disclosed herein for generating augmented training examples to fine-tune a machine learning model for converting natural language to logical form. In one aspect, a method is disclosed that includes: accessing a set of training examples, wherein each training example includes: (i) an original prompt having a database schema and (ii) a programming language completion corresponding to the original prompt, generating a set of augmented training examples based on the set of training examples, generating a fine-tuning dataset comprising the training examples and the augmented training examples, and fine-tuning a model with the fine-tuning dataset. Generating the set of augmented training examples can include performing an iterative process including: selecting a subset of database object names from the database schema and/or programming language completions and performing transformations to each database object name of the selected subset of database object names within the original prompt and programming language completion.

SYSTEM AND METHOD FOR SMART ELECTRICAL LOAD MANAGEMENT

Resumen de: WO2026198742A1

A system and method of managing an electrical power grid. The method comprises: obtaining measurement data comprising a respective time series for each of a plurality of electrical loads connected to an electrical power grid, each time series comprising electrical power consumption measurements for the corresponding electrical load at each of a plurality of times; processing each of the time series using a machine learning model to determine a respective classification output for each of the electrical loads; processing the measurement data to predict one or more states of the electrical power grid at one or more corresponding times; and adjusting electrical power distribution to one or more of the electrical loads based on the classification outputs for the electrical loads and the predicted one or more states of the electrical power grid.

UTILIZING ARTIFICIAL INTELLIGENCE FOR DATA MANAGEMENT AND DATA VISUALIZATION

Resumen de: US20260288716A1

A first dashboard is generated utilizing a machine learning model and user settings associated with a user. One or more follow-up prompts are received. Data needed to answer the one or more follow-up prompts is determined and obtained utilizing one or more data catalogs. One or more subsequent dashboards are generated utilizing the machine learning model and the user settings associated with the user.

MODEL DEPLOYMENT SYSTEM FOR GENERATING OPTIMIZED MODELS FOR A TARGET ENVIRONMENT

Resumen de: US20260288437A1

A model deployment system (MDS) is disclosed that includes capabilities for generating an optimized model that is tailored to specific target deployment hardware used to deploy the model in a target environment. The MDS is capable of executing a model inference accelerator (MIA) layer to perform a set of optimizations on a machine learning (ML) model. The MIA layer executes the set of optimizations based on model metadata associated with the ML model to generate an optimized ML model. The execution performed by the MIA layer includes generating an intermediate representation (IR) format for the IR model that identifies model-specific optimizations to be applied to the ML model and target hardware for deploying the ML model. The execution by the MIA layer further includes compiling the IR format of the ML model based on model-specific optimizations and target hardware-specific model optimizations to generate an optimized ML model.

GENERATING SPATIALLY DISTRIBUTED RESPONSE TO QUERY BY INTELLIGENT SELECTION OF TIME-SERIES REMOTE SENSING DATA

Resumen de: US20260288888A1

This disclosure relates generally to a method and system for generating spatially distributed response to query by intelligent selection of time-series remote sensing data. The current available methods for processing and analyzing geospatial data have limitations such as manual satellite data selection, limited natural language query support and the like. The method disclosed receives multi-contextual user query, intelligently selects the relevant time-series satellite data and climate information from repository using intents and entities generated utilizing pre-defined knowledge graph. The method intelligently selects time-series satellite data by creating and utilizing relationship strength-based node ranking graph. This time-series satellite data is combined with the climate data selected from climate repository to generate the spatially distributed response for the multi-contextual user query using machine learning model and large language model. The method enables stakeholders to access actionable insights without needing expertise in satellite data processing and using it for informed agricultural decision-making.

METHODS AND SYSTEMS FOR USING CONTENT CONTRIBUTION VALUE TO DETERMINE MEMORY, STORAGE, CONTEXT WINDOW, AND MODEL PARAMETERS IN MACHINE LEARNING MODELS

Resumen de: US20260288521A1

Systems, methods, and computer-readable media determine allocation of model memory, storage, parameters, context window capacity, tokens, or training and inference resources based on contribution values of input data. A system can encode input data, segment the input data into input sequences, test a model with the input sequences, obtain output token or encoding sequences and logit values, and compute entropy, significance, weight of evidence, explanatory power, rarity, or salience to determine expected contribution values. Input sequences having low or negative contribution value can be discarded or reordered, and remaining input sequences can be retained for training, inference, fine tuning, reinforcement learning, or context-window assembly. The resulting allocation can reduce processing of low-utility data while optimizing memory, storage, and parameter resources of the model in training.

SYSTEM

Resumen de: US20260290380A1

A system includes a processor that is configured to extract, from a database, data related to past business performance and customer traffic for a corresponding past period and analyze the extracted data using a statistical analysis technique, generate a prompt for instructing a generative artificial intelligence model to generate advice, input the prompt into the generative artificial intelligence model, and obtain advice generated by the generative artificial intelligence model, and perform preprocessing on data using an information processing device, normalize the data, and perform feature extraction on the data so as to apply a machine learning algorithm and construct a prediction model.

AUTONOMOUS MULTIFACTOR GENERATIVE ARTIFICIAL INTELLIGENCE FRAMEWORK

Resumen de: US20260289633A1

0000 Various embodiments are directed to apparatuses, methods, computer-readable media, computer program products, and systems related to detecting a process trigger that identifies a target entity; identifying, entity data associated with the target entity; generating, using a machine learning based prediction model, a predictive asset output based at least in part on the entity data, wherein the predictive asset output comprises at least one predicted asset; generating, using a dynamic contextualization model, a multifactor contextualized asset representation for the at least one predicted asset based at least in part on the at least one predicted asset and the entity data; and transmitting the multifactor contextualized asset representation to one or more computing devices via one or more communication channels.

METHOD OF MODEL TRAINING, ELECTRONIC DEVICE AND STORAGE MEDIUM

Resumen de: US20260289351A1

Provided is a method of model training, an electronic device and a storage medium, relating to the field of data processing technologies, and in particular, to the technical fields of artificial intelligence, big data, large models and deep learning. The method includes: obtaining target sample data corresponding to N branches required in a current training period of distributed training; performing T update operations on network parameters of a preset model deployed on the N training nodes for the distributed training based on T data blocks corresponding to the branches to obtain first full parameter sets corresponding to the branches; performing model parameter fusion on the first full parameter sets to obtain a fused full parameter set; and updating the network parameters of the preset model deployed on the training nodes based on the fused full parameter set and proceeding to the next training period to obtain a target model.

SYSTEM

Resumen de: US20260289449A1

0000 A system includes a processor that is configured to collect information for attracting visitors to a facility from at least one sensor or at least one database and analyze the collected information by using a machine learning algorithm, acquire attributes of visitors by using at least one camera or at least one RFID sensor and analyze the acquired attributes by using a data mining technique, and generate, by using a natural language processing technique, a prompt sentence for instructing a generative artificial intelligence model to generate event content and input the prompt sentence to the generative artificial intelligence model.

METHOD FOR SECURELY DEPLOYING A MODEL, DEVICE AND STORAGE MEDIUM

Resumen de: US20260289400A1

A method for securely deploying a model, a device, and a storage medium are provided. The method includes: receiving, from a user side, a deployment request for a machine learning model, the deployment request including at least a model identification of the machine learning model that is encrypted and a key identification of a key used to encrypt the machine learning model; obtaining the machine learning model that is encrypted and the key based on the deployment request; decrypting, in a trusted execution environment, the machine learning model that is encrypted using the key to obtain the machine learning model; and deploying the machine learning model in the trusted execution environment.

INFORMATION PROCESSING METHOD AND RELATED DEVICE

Resumen de: US20260285373A1

0000 Embodiments of this disclosure disclose an information processing method and a related device. The method may be applied to the autonomous driving field in artificial intelligence. The method includes: inputting, to a deep learning model, first information corresponding to a traffic scene around an ego vehicle, and obtaining second information corresponding to the first information, where the first information includes first text information, the second information is obtained based on the deep learning model, and the second information corresponds to any one of the following tasks: making a decision on behavior of the ego vehicle, planning a trajectory for the ego vehicle, or controlling the ego vehicle. Input information of the deep learning model provided in this disclosure includes text information that facilitates understanding by a user.

SENTENCE GENERATION MODEL GENERATION DEVICE, SENTENCE GENERATION MODEL GENERATION METHOD, AND SENTENCE GENERATION DEVICE

Resumen de: US20260289073A1

The sentence generation model generation device (10) includes a learning unit (11) configured to, in response to input of an input sentence, identify, from the input sentence, a claim and a premise that supports the claim from the same standpoint as the claim or from a different standpoint than the claim, perform at least one of labeling the claim and the premise and labeling a relation between the claim and the premise, and conduct machine learning using the input sentence as an explanatory variable and the text of the labeled portion as an objective variable, to perform training of a sentence generation model for generating and outputting a text sentence related to a structure of the input sentence and generate the sentence generation model.

DYNAMICALLY MONITORING VARIATIONS IN PROCESS EXPLAINABILITY WITHIN A DISTRIBUTED COMPUTING ENVIRONMENT

Resumen de: US20260289340A1

The disclosed embodiments include computer-implemented systems and processes that dynamically monitor variations in process explainability of within a distributed computing environment. For example, an apparatus may determine a first metric value characterizing a variation in an explainability of a machine-learning or artificial-intelligence process between a first temporal interval and a second temporal interval. When the first metric value is inconsistent with an exception criterion, the apparatus may obtain a second metric value characterizing an additional variation in the explainability of the machine-learning or artificial-intelligence process between the first temporal interval and a third temporal interval, and when the second metric value is inconsistent with the exception criterion, the apparatus may perform operations that modify at least one of (i) a value of a process parameter that characterizes the machine-learning or artificial-intelligence process or (ii) a composition of an input dataset associated with the machine-learning or artificial intelligence process.

IDENTIFICATION OF MARKER SEQUENCES FOR RAPID CLASSIFICATION OF MICROBIAL PATHOGENS THROUGH MACHINE LEARNING ANALYSIS OF GENOME ASSEMBLIES

Resumen de: US20260290509A1

0000 The disclosure provides methods to identify marker sequences for rapid classification of microbial strains through machine learning analysis of genome assemblies.

APPARATUS AND METHOD FOR IDENTIFYING BASE STATION BEHAVIOUR

Resumen de: US20260292508A1

0000 A method for identifying anomalous base station behaviour in a communications network comprises generating an ensemble algorithm ML model, performing first training of plural decision trees in the ML model and performing second training of the decision trees such that feature values of diversion nodes are included in feature contributions of anomaly detection nodes. The method further includes processing communications network report data using the ML model, identifying a base station exhibiting anomalous behaviour and generating an alert. A machine learning agent, a communications network, a computer program and a computer-readable storage medium are also disclosed.

APPARATUSES AND METHODS FOR MACHINE LEARNING MODEL RELIABILITY ASSESSMENT

Resumen de: US20260289336A1

Example embodiments enable reliability assessment of a machine learning model for mobility management related predictions. A client node may be configured to obtain configuration information for reliability assessment of a machine learning model, the configuration information comprising one or more parameters for configuration of an evaluation window, one or more evaluation functions and one or more reliability acceptance thresholds; obtain verification data based on the evaluation window and evaluation criteria for determining at least one reliability measure; determine at least one reliability measure based on the verification data and the one or more evaluation functions; determine qualification of the machine learning model based on the at least one reliability measure and the one or more reliability acceptance thresholds; send a result of the qualification to the network node; receive feedback from the network node based on the result indicating if the client node is authorized to use the machine learning model; and perform a mobility management related prediction based on the machine learning model when authorized by the network node. Apparatuses and methods are disclosed.

METHODS AND SYSTEMS FOR CONCRETE MATERIALS AND CONCRETE MIXTURE CHARACTERIZATION

Resumen de: US20260289427A1

Systems, methods, and computer-readable media of characterizing concrete materials or concrete mixtures may include: obtaining, using sensors that comprise at least two different types of sensors, sensor data corresponding to the concrete materials or the concrete mixtures; analyzing, using a trained machine learning model, the sensor data to generate the characterization of the concrete materials or the concrete mixture; and outputting the characterization of the concrete materials or the concrete mixture. Training the trained machine learning model may include: obtaining a set of training data for historical concrete materials or a historical concrete mixture, including a plurality of characterizations of the historical concrete materials or the historical concrete mixture; classifying the set of training data into a plurality of subsets each corresponding to a different characterization or range of characterizations of the plurality of characterizations; and generating the trained machine learning model using the plurality of subsets of training data.

SYMBOLIC MACHINE LEARNING ARCHITECTURE

Resumen de: US20260289401A1

A machine learning system can include a symbolic representation generator and an inference component. The symbolic representation can generate a symbolic memory network using input obtained from an environment. The inference component can generate a causal model by experimenting upon components, operations, and parameters of the symbolic representation generator. The causal model can influence the generation of the symbolic memory network. In particular, the machine learning system can identify an analog in the symbolic memory network as being entangled, identify a disentangled version of the identified entangled analog in the symbolic memory network, and, in response to the identification of the disentangled version of the identified entangled analog, release the identified entangled analog from the symbolic memory network and provide instructions to take an action in an environment based on the disentangled version of the identified entangled analog.

SYSTEM AND METHOD FOR MATCHING ENTITIES USING MACHINE LEARNING

Resumen de: US20260288480A1

0000 A system and method may identify the best matching entities using artificial intelligence, by for example determining, from a set of features describing a number of entities, a subset of features which when input to a machine learning module allow the machine learning module to match a pair of entities based on the ability of the second entity in the pair to successfully take over responsibilities handled by the first entity in the pair; training a machine learning module using the determined features; and for a first entity described by the determined set of features, using the machine learning module to determine a list of entities matching the first entity, each entity in the list of entities associated with a matching score.

SMART RING SYSTEM FOR MONITORING SLEEP PATTERNS AND USING MACHINE LEARNING TECHNIQUES TO PREDICT HIGH RISK DRIVING BEHAVIOR

Resumen de: US20260285376A1

A method can include acquiring, by a sensor of a sleep detecting device, a set of user data indicative of sleep activity of a user. The method further can include acquiring data indicative of an activity of the user. The method also can include predicting, based at least on the set of user data and the activity of the user, a level of risk exposure for the activity of the user. The method also can include in response to predicting, generating an output indicative of the level of risk exposure of the user. Other embodiments are disclosed herein.

METHOD FOR ESTABLISHING ROBOT COMMUNITY, ROBOT COMMUNITY SYSTEM AND OPERATING METHOD

Resumen de: EP4811249A1

0001 A method for establishing a robot community, a robot community system and an operating method. For establishing the robot community, each of intelligent robots (101) is assigned with a specific attribute, and accordingly a training data is prepared for training the intelligent robot (101) through a machine-learning algorithm. The robot communication system then deploys multiple intelligent robots (101) into a social media and generates community contents associated with each of the intelligent robots (101) according to the attribute and basic data of each of the intelligent robots (101) by a generative intelligent model. Therefore, the robot community system allows users to interact with the intelligent robots (101) and establish social relationships between the users and the intelligent robots (101) in the social media.

QUANTUM COMPUTING FOR ACCELERATING MACHINE LEARNING MODEL TRAINING

Resumen de: WO2026075653A2

A system and method to accelerate machine learning training comprises receiving, by a non-quantum computer, a set of data for training a machine learning model; dividing the set of data into a plurality of data batches, each data batch determined according to predetermined requirements of a quantum computer; calculating, by the non-quantum computer, a Jacobian matrix for each of the plurality of data batches, the Jacobian matrix and a loss vector being part of a linear matrix equation for updating weights in the model; arranging, by the non-quantum computer, input data of a polynomial function formatted for a quantum computer; solving, by the quantum computer, a vector of weight updates; and updating the weights in the model.

MACHINE LEARNING BASED METHOD FOR CONSIDERING CONSTRAINT VIOLATIONS WHEN SELECTING OPTIMAL ACTIONS TO PERFORM IN A COMMUNICATIONS NETWORK

Resumen de: WO2025105999A1

Embodiments described herein relate to methods and apparatuses for initiating control of one or more parameters in a communications network. A computer-implemented method comprises obtaining a first state of the communications network; for each action in a set of actions: predicting a next state of the communications network utilizing a first ML model based on the first state and the action; predicting, utilizing a second ML model, whether taking the action whilst the communications network is in the first state will violate one or more constraints; determining a cost value associated with the action by evaluating a cost function, wherein the cost function comprises: a first component representing one or more performance metrics of the communications network in the next state; and a second component dependent on whether taking the action in the first state will violate the one or more constraints; selecting a selected action from the set of actions, wherein the selected action is associated with a best cost value; and initiating performance of the selected action in the communications network.

SYSTEM AND METHOD FOR MONITORING THE OPERATING CONDITION OF ROTARY ELECTRICAL MACHINES AND AUTOMATIC DETECTION OF MECHANICAL AND ELECTRICAL FAULTS

Nº publicación: EP4811625A1 23/09/2026

Solicitante:

2NEURON SOLUCOES EM INTELIGENCIA ARTIFICIAL LTDA [BR]
2Neuron Solucoes Em Inteligencia Artificial Ltda

Resumen de: EP4811625A1

The present invention pertains to methods for operation, maintenance, and monitoring of rotating electrical machines, and relates to a system and a method for monitoring the operational condition and detecting mechanical, electrical, load, and process faults in rotating electrical machines. The proposed system and method, together, provide a more effective way to detect faults early on, with greater installation convenience and scalability than prior art methods, and are more efficient in avoiding production losses, improving operational performance and preventing damage to equipment and risks to operators. The developed method continuously collects electrical current and voltage signals that power the machine, utilizing a data acquisition module and current and voltage transformers. The collected data undergoes stages of compression, encryption, application of Fast Fourier Transform (FFT), subsampling, feature extraction, anomaly detection using statistical and machine learning techniques, and fault classification using machine learning techniques. The proposed system consists of one or more data acquisition modules, one or more gateway devices, a processing center on a cloud computing platform, and an operator interface. In an industrial application, the method can be continually improved with the collection of new data and with validation information from an operator in the face of an identified fault.