2070results about "Natural language analysis" patented technology

A system may be configured for cooperation between language model agents. An agent may be, for example, a computer system, or a software component executing on a computer system, that can accept text and / or natural language inputs, draw upon an LM to process the inputs and perform a function, and respond via text and / or natural language outputs. An agent may act as a mediator to interact with a user, identify a task requested by the user, and delegate one or more subtasks to another agent or other resource. An agent may act as a delegate to handle tasks or subtasks delegated by a mediator. Agents may communicate with each other using a combination of structured and unstructured language; for example, one or more parameters and a natural language message.

Providing generative artificial intelligence (AI)-enabled notebook interfaces for a security framework, including: receiving a request to generate a notebook interface for a security framework monitoring a cloud deployment; generating, in response to the request, the notebook interface, wherein the notebook interface comprises one or more notebook cells for interacting with the security framework, wherein the one or more notebook cells comprise a natural language input cell for querying a generative artificial intelligence (AI) model; and presenting the notebook interface.

Techniques are described for performing automated operations related to generating and providing housing-related information, such as to automatically respond to free-form natural language query requests for housing-related information of various types received by a chatbot by using a combination of automated tools to generate and provide responsive housing-related information. In at least some situations, the described techniques generates responses using a trained large language model that maintains context over an interaction session with a user involving multiple user queries and corresponding responses, and ensuring accurate response information by restricting the generation of the response information in particular ways and by identifying and providing citations to authoritative sources used to generate the response information.

A method for training a machine learning (ML) model using a large language model (LLM) is provided. A system for detecting fraud which utilizes the LLM-trained ML model trained is also provided. An artificial intelligence (AI)-based method for monitoring alerts is also provided. The method for training an ML model using an LLM includes receiving tabular data for training the ML model, generating one or more natural-language strings comprising information from the tabular data, generating, via a base LLM, one or more prompts and completions based on the one or more generated natural-language strings, pre-training the base LLM using a plurality of generated prompts and completions, updating the base LLM via supervised learning using a cross-entropy loss function with ground-truth labels, and fine-tuning the updated LLM via reinforcement learning with human feedback using a reward model and a proximal policy optimization model to produce the LLM-trained ML model.

A document preparation method involves creating a hierarchical representation of an input document without summarizing or omitting any content. The method uses a generative language model to generate the hierarchical representation and stores it in a repository for later use by a client generative language model. This allows for more accurate and complete generation of text, enabling the use of retrieval units to enhance the output of the client generative language model while efficiently exploiting its limited context window.

The invention belongs to the technical field of intelligent information retrieval, and provides a multi-factor value evaluation method and device for intelligence resource optimal configuration, and the method comprises the steps: obtaining a query context corresponding to an intelligence auxiliary decision-making task; in response to the query context, scoring the intelligence to be evaluated in parallel on a plurality of intelligence value scoring factors to obtain factor scores corresponding to the intelligence value scoring factors; performing confidence coefficient evaluation on each factor score to obtain a corresponding confidence coefficient; according to the confidence coefficient of each factor score, correcting the corresponding basic weight to obtain a corresponding corrected weight; and according to all the factor scores and the correction weights thereof, obtaining a comprehensive score of the to-be-evaluated intelligence, the comprehensive score being used for performing priority ranking on the to-be-evaluated intelligence.

Systems and methods for domain-specific model-generated content item generation, evaluation, and selection can include generating a plurality of candidate model-generated content items that can then be evaluated based on one or more signals, which can then be leveraged for candidate model-generated content item selection. The plurality of candidate model-generated content items can be generated with a generative model that was tuned for domain-specific content item generation. The selected model-generated content item can be processed to generate an outline that may then be provided to a user for user interaction to generate an augmented outline. The augmented outline may then be processed to generate an updated model-generated content item.

A system and method for real-time or near real-time multilingual language transformation in telecommunications environments using distributed artificial intelligence (Al). The system includes at least one processor configured to instantiate a plurality of Al agents, each corresponding to a user in a communication session, and to manage their operation via an Al-agent runtime adapter comprising a microkernel agent manager and a virtual machine layer. A language model abstraction layer enables access to language models and Al algorithm libraries through publish-subscribe interfaces. The system supports contextual transformation of speech based on user profile data such as sentiment, emotional tone, and cultural background. Execution can occur on telecommunication-class switches, edge devices, or general-purpose computing hardware. Al agents operate asynchronously, access shared or private memory, and produce personalized, bidirectional linguistic output. The architecture supports secure, scalable deployment across heterogeneous devices and networks.

A store employee assistance platform is provided that is enabled with generative artificial intelligence to deliver role-specific, context-aware responses to natural language questions submitted by users, including store employees. The platform includes a store employee assistance application with a chat interface through which a user may submit a question. The platform identifies the user's role, store location, and access level, and retrieves relevant enterprise content from a vector database and historical data store. A prompt engine constructs a contextualized prompt incorporating the user's identity and the retrieved information, and submits the prompt to one or more generative AI models. The resulting response is tailored to the user's responsibilities and delivered through the chat interface, providing real-time operational guidance specific to the user's role within the retail environment.

An apparatus for generating obfuscated data within a computing environment, comprising a processor and a memory containing instructions configuring the processor to access a database containing a plurality of private data elements belonging to at least a private record, generate a set of obfuscated data elements, representative of the at least a private record, as a function of the plurality of private data elements using an generative model, determine a first distance measure between at least an obfuscated data element within the set of obfuscated data elements and at least a private data element of the plurality of private data elements within the database, and verify the first distance measure is within a distance range, wherein a minimum threshold of the distance range is determined as a function of a deidentification parameter and a maximum threshold of the distance range is determined as a function of an obfuscation parameter.

There are provided systems and methods for a domain-knowledge guided agent framework for automated system analysis. An online transaction processor or other service provider may provide computing services and platforms to entities, which may require compliance enforcement for different policies, regulations, and the like. To provide compliance review, investigations, and enforcement in a computing system of a service provider, the service provider may implement an intelligent and automated agent and framework that may utilize different large language models for processing compliance investigation requests and queries. The agent may utilize the models to plan and execute tasks using an available toolkit of computing operations and capabilities for compliance investigation. A data guard module may also be used to ensure data privacy and security is maintained. Within a main task, sub-tasks may be executed by models with specific domain knowledge.

A Large Language Model (LLM) is configured to automatically label non-labeled textual data-items for the purpose of creating a training dataset for training a Machine Learning (ML) model. The ML model is thus trained on LLM-labeled textual data-items; and the ML model can be deployed to classify new or incoming documents or messages or other textual data-items. Additionally, a Vision and Language Model (VLM) or a Large Multimodal Model (LMM) or a large multiple-modalities model (LMM) can process data from two or more modalities (visual data, textual data), and is configured to automatically label non-labeled images for the purpose of creating a training dataset for training a Deep Neural Network (DNN) or a Deep Convolutional Neural Network (Deep CNN) model. The DNN model is thus trained on VLM-labeled images; and the DNN model can be deployed to classify new or incoming images.

This disclosure relates to techniques for generating and utilizing custom stickers in a digital communication environment. A technique involves receiving a text-based message input during a chat session and using a generative language model (e.g., a Large Language Model, or LLM) to create a text prompt. This prompt is then used by a generative image model to produce a custom sticker. The generated sticker is sent to a client device where it is displayed in a sticker tray alongside other selectable stickers. Users can select and send these stickers directly within their chat interface, enriching communication with visually expressive and contextually relevant imagery.

A robotic controller for controlling a robot according to a sequence of robotic actions. comprises an input interface configured to receive a plurality of multimodal inputs each specifying instructions for performing a task in a different modality including audio, video, and a text modality. The controller also comprises a multimodal large language model, an action sequence decoder, and a controller. The multimodal LLM includes a multimodal LLM encoder and an LLM decoder. The multimodal LLM encoder is trained with machine learning to transform the multimodal instructions into encodings and the LLM decoder is configured to decode the encodings into a sequence of robotic instructions. The action sequence decoder is trained with machine learning to transform the sequence of robotic instructions into a sequence of actions using a library of robotic skills. The controller is configured to control a robot according to the sequence of actions.

Systems, methods, and devices for monitoring and controlling communications between autonomous agents are disclosed. The system monitors real-time communications between autonomous agents, intercepting and recording each communication. Communications are translated to a standardized language and processed through communication protocol filters that evaluate compliance with predefined operational policies. The system parses each translated communication to identify policy violations. When violations are detected, the system modifies communications to ensure compliance with operational policies. All communications and modifications are recorded via distributed ledger technology for audit and accountability purposes. This approach enables comprehensive oversight of autonomous agent interactions while maintaining tamper-proof records of all monitoring and control activities.

There is provided a method of automatically detecting that a target image is deepfake, comprising: receiving authentic images depicting a vehicle with actual damage, receiving the target image depicting potential damage to the vehicle, feeding the target image into a machine learning (ML) model, obtaining a candidate set of human-readable text describing the potential damage to the vehicle, feeding the authentic images into the ML model, obtaining from the ML model, a ground truth set of human-readable text describing the actual damage to the vehicle depicted in the authentic images, computing a similarity metric indicating a difference between the potential damage described in the candidate set of human-readable text and the actual damage described in the ground truth set of human-readable text, and in response to the difference being above a threshold or meeting a requirement indicating a significant difference, detecting that the target image is likely deepfake.

Various embodiments discussed herein relate to route optimization and query understanding for route and / or direction queries with complex user preferences. Each route candidate, for example, is treated as a richly annotated document. The routing engine, in addition to performing route optimization, acts as a retriever and ranker of route documents according to user intent. Various embodiments rank routes not just based on a simple cost model, but based on many more or alternative factors according to user preferences, user intent, and / or contextual data.

The systems and methods disclosed herein receive alphanumeric characters defining operative boundaries for expected model use cases, along with operational data. The expected model use cases share common attributes, which are used by a first AI model to construct observed model use cases from the operational data. Each observed model use case includes features such as a text-based description, expected input and output,? AI model(s) generating the expected output from the input, and / or data supporting the AI models. For each observed model use case, a second AI model maps the alphanumeric characters and features to a risk category, selecting from multiple risk categories based on the level of risk associated with the features. The system identifies criteria for the observed model use case within the alphanumeric characters and generates gaps by comparing the criteria with the features of the observed model use case.

A method for automated control code generation and verification includes: receiving a natural language command, the natural language command being configured to instruct the large language model to output a code text that meets control requirements corresponding to the natural language command; performing matching retrieval on a vector database according to the natural language command to obtain a sample code snippet; obtaining API structured information corresponding to the sample code snippet from a knowledge graph database; generating an initial control code according to the sample code snippet and the API structured information; and performing a multi-level virtual operation verification on the initial control code in a software motion control system, and generating a target control code according to multi-level verification results confirmed multiple times by the user and the initial control code.

Systems and methods are provided to improve data retrieval from a cache memory by using semantic matching to retrieve data from the cache memory. The system includes a two-tiered cache system, with a first tier implementing “key-value” pairs, and a second tier that includes a table that is configured as an artificial intelligence (AI) search indexed source. When a new input does not have a matching “key” at the first tier, the system performs a semantic search at the second tier of the cache to determine if relevant data is stored in the cache. The current systems and methods increase the likelihood of obtaining data for queries from the cache memory, reduce the response time to the queries, improve search consistency, reduce computing resource utilization, improve system performance, and reduce costs.

An approach is provided for large language model (LLM) map feedback reporting. The approach involves, for example, processing an input specifying a map error / feedback using an LLM to classify a map error type. The approach also comprises determining a map error template based on the map error type that specifies structured data fields for the map error type. The approach further involves using the template to construct prompts for the LLM to generate questions to collect data items for populating the data fields, and providing the prompts to the LLM to generate the questions and collect the data items from the user. The approach further involves using template to generate additional prompts to generate a map error report of the data items in a structured data format, and providing the additional prompts to the LLM to generate the map error report for transmission to a map feedback system.

The invention provides an intelligent customer service interaction content recommendation method and system based on artificial intelligence, and the method comprises the steps: obtaining power grid equipment state data and user behavior data, and carrying out the weighted fusion through an attention mechanism, and outputting a fusion feature vector; predicting a prediction vector set of a future time period based on the fused feature vector; screening a demand event set higher than a threshold in the prediction vector set, and generating a recommended content set; performing emotion recognition according to a current input text of the user, and calculating a recommended content score and a pushing priority score in combination with the recommended content set; a cross-modal consistency regular term is introduced to correct the push strategy vector, and an adjusted recommendation push score and a corresponding recommendation content set are obtained; and converting the adjusted recommendation push score and the corresponding recommendation content set into a scheduling task, and performing linkage execution with a scheduling system. According to the invention, efficient, active and intelligent upgrading of the intelligent customer service system in a power grid scene is realized.

Systems and processes enhance cybersecurity by dynamically generating a decoy sub-application that operates parallel to the primary application using generative design and a large language model. The core feature involves real-time anomaly detection within application traffic, utilizing AI to assess threats and orchestrate appropriate responses. Upon identifying potential security threats, the system employs generative design principles to architect a restricted-functionality sub-application, deploying it instantly to engage and analyze the attack vectors without compromising sensitive production data. The sub-application is isolated through software-defined networking, ensuring that its operations do not affect the primary application's functionality. Additionally, sophisticated traffic redirection mechanisms are employed to divert suspicious traffic from the primary to the decoy application, thereby protecting the integrity while allowing detailed threat analysis. This dual-capability system not only safeguards against disruptions but also enhances adaptive security measures through continuous learning and system adjustments.

The invention discloses an engineering drawing compliance intelligent review method and system constructed based on a knowledge base and a large model. The method mainly comprises the following steps: intelligently analyzing a design specification in a natural language form by utilizing a large language model, and dynamically constructing a machine-readable rule base and a knowledge base; uniformly converting the multi-format engineering drawing into a structured intermediate format; functional areas and primitives in the drawing are recognized through a computer vision model, and natural language information describing attributes of the functional areas and the primitives is generated; and finally, in combination with the knowledge base and natural language description, performing compliance judgment by utilizing the reasoning ability of the large language model, and generating an interpretable review report. According to the method, computer vision and a large language model technology are fused, so that the defects that traditional manual review is low in efficiency and prone to making mistakes and a traditional automatic tool lacks semantic understanding ability are overcome, and efficient and accurate engineering drawing automatic compliance review with deep semantic understanding ability is achieved. Figure 1 of the abstract is a system architecture block diagram.

The invention discloses a document auditing method and device based on cooperation of a large model and a rule engine. The method comprises the following steps: analyzing an original document to form an intermediate representation file; extracting entities from the intermediate representation file to form an entity candidate set; inputting the thinking chain cue word into the large model to obtain a first triple, a first confidence coefficient and a first risk level, and calculating a first traceable score of the large model according to the reasoning path node; performing deterministic judgment by using the rule engine to obtain a second triple, a second confidence coefficient, a risk level and a second traceable score; calculating two weighted voting values and obtaining a conflict difference value; taking a conclusion corresponding to the high-weighted voting value as a final conclusion when the conflict difference value is smaller than a preset judgment threshold value; otherwise, starting an artificial rechecking process; and generating audit reports in various formats. According to the method, the advantages of relatively high language understanding ability of a large model and relatively high certainty of a rule engine are exerted, and the problem of missed checking or excessive marking is avoided.

This application describes, amongst other things, methods and systems for building and deploying agents. An example method includes obtaining orchestration data about a set of task-specific components selected to provide a response to the prompt, where each respective task-specific components in the set of task-specific components is configured to assist with a respective clinical task of the one or more clinical tasks. The method further includes, determining an order in which each respective task-specific components of the set of task-specific components should be utilized to prepare a complete response to the prompt that address the one or more clinical tasks based on the obtained orchestration data about the set of task-specific components. The method also includes, in accordance with the determined order, providing first data related to the prompt to a first task-specific component and receiving a first response from the first task-specific component.

Techniques for evaluating the output of a large language model are disclosed. A training data set that includes deterministic computational metrics that measure features of large language model output and qualitative metrics that provide a non-deterministic measure of large language model output quality may be used to train a ML model. The ML model may then be used to estimate the qualitative metrics of large language model output by using deterministic computational metrics as input.

Examples of the present disclosure describe systems and methods for automating the identification of events in a text file. In examples, a computing system identifies a subset of a text file that comprises an unknown event using a set of rules. Each rule of the set of rules specifying a first pattern of characters is compared to the subset of the first text file. When the set of rules does not identify the unknown event, the subset of the text file is provided to a language model to generate a new rule with a second pattern of characters and an identifier of the new rule. The system then generates an updated set of rules by adding the new rule to the set of rules.

Provided is a cyber threat information processing method including receiving a CTI analysis request for assembly code from a client; analyzing the assembly code to obtain analysis information of the CTI for the assembly code; generating a CTI query related to a file based on the analyzed CTI and delivering the CTI query to a natural language model; and providing natural language description information according to the CTI query obtained from the CTI for the assembly code and the natural language model.

A method for processing cross-modal question answering based on large model, an apparatus, and a storage medium are suggested, which relates to the field of artificial intelligence technologies such as speech interaction processing, large models, machine learning and natural language processing. The specific implementation includes: performing an activity detection on a target speech input by a user; in response to detecting a pause in the inputting of the target speech, obtaining a first text corresponding to a first input speech before the moment of the pause in the target speech; performing a text response processing using a pre-trained speech question answering processing system based on the first text and the first input speech.