470 results about "Response generation" patented technology

A method for using a local large language model (LLM) within a user's secure computing environment is disclosed. The LLM operates behind a firewall to prevent transmission of sensitive data, and utilizes an encrypted vector database and artificial intelligence techniques for content retrieval, response generation, and task anticipation. This system can be used on mobile, wearable, vehicle, or IoT devices and offers various services such as health monitoring, financial advice, automated communications handling, and personalized daily activity optimization. It also has the ability to detect fraud, fine-tune responses using augmented user data, assist in negotiations, identify personal interests, and provide health recommendations based on dietary and physical activity data.

The invention discloses a tourism service agent system and method based on a multi-modal large model. The system comprises five parts: a user interaction interface, which is used for receiving a user request and returning a response; the agent center is used for performing intention understanding, task planning and response generation through a multi-modal large language model; and the tool calling module is used for executing specific API business operation. According to the intelligent tourism service system, the work of each module is coordinated through the intelligent agent center, a complete closed loop from user intention understanding to service execution is realized, accurate, reliable, whole-course and personalized intelligent tourism service can be provided, and the problems that a traditional tourism service system is single in function, inaccurate in information and lack of action ability are effectively solved.

Disclosed herein, is a method and system for selecting a LLM for response generation to user queries. The method includes receiving a user query from a user device. The method includes determining, for the user query, a query type from a set of query types through a fine-tuned text classification model. The method includes retrieving a plurality of document embeddings based on the user query and the query type from a vector database through a semantic search technique. The method includes preparing a prompt using the user query and the plurality of document embeddings. The method includes inputting the prompt to an LLM selected from a set of LLMs based on the query type. The method includes generating, via the selected LLM, a response to the user query based on the prompt.

There is provided a computer method, system and device comprising detecting an updated iteration of a document for query response generation; comparing the updated iteration to a prior iteration to identify chunks of the updated iteration of the document that differ from corresponding chunks of the prior iteration, the prior iteration for generating a set of synthetic questions and answers using an LLM. Responsive to identifying that a given chunk of the updated iteration differs from a corresponding chunk of the prior iteration, the method triggers generation, using the LLM, of a new set of synthetic questions associated with corresponding text in the given chunk defining a new set of synthetic responses, and wherein the new set of synthetic questions and responses replaces at least a subset of the set of synthetic questions and answers associated together by a mapping with the corresponding chunk of the prior iteration.

The invention discloses a multi-context semantic recognition and understanding method based on a large language model. The method comprises the following steps: S1, generating a semantic unit sequence; s2, constructing a context nested vector sequence; s3, constructing context priori representation, and generating a semantic representation sequence; s4, outputting a state vector of the semantic path by adopting a gating loop unit, obtaining a matching degree score according to a feedforward neural network, and determining a deliberate map tag and an alternative intention tag; s5, slot field extraction and semantic filling are completed, and a structured semantic task unit is generated; and S6, completing semantic recognition and service response closed loop. According to the method, by introducing a prefix regulation and control mechanism and a multi-context semantic modeling structure, the accuracy of intention recognition in multiple rounds of dialogues and the consistency of the context generated in response are remarkably improved, and the method is suitable for natural language understanding scenes of multi-language intelligent customer service, cross-context man-machine interaction and complex task driving.

The invention is suitable for the technical field of natural language processing, relates to a hierarchical memory and context awareness retrieval method of a large role model and a related product, and aims to solve the problems of limited model memory duration, insufficient retrieval correlation and insufficient personality consistency in a long dialogue. According to the invention, a short-term-middle-term-long-term three-level memory architecture is adopted, and a memory attenuation and migration mechanism is combined, so that dynamic metabolism of memory is realized; related memories are recalled accurately through a context semantics and role personality double-sensitive double-stage retrieval algorithm; relying on a personality-linked memory fusion and response generation strategy, the reply is ensured to fit personality setting; and a closed-loop adaptive learning mechanism of dialogue-memory-retrieval-generation-feedback is constructed, and the memory quality is continuously optimized. According to the method, the role large model can have the human-like continuous memory ability, the continuity, retrieval accuracy and personality consistency of long dialogues are remarkably improved, and the long-term personalized interaction requirements of scenes such as digital personality assistants and dialogue agents are met.

The invention relates to the technical field of data analysis, can be applied to agricultural disaster monitoring, financial science and technology and other business scenes, and discloses a multi-source monitoring analysis and decision-making method, device, equipment and medium, and the method comprises the steps: obtaining multi-source monitoring data, and carrying out the time-space multi-scale fusion to generate time-space fusion data, extracting difference features of the target object before and after the event based on the fusion data to identify a first type of events, predicting a second type of events based on dynamic time sequence features by using a space-time diagram network model, and constructing a knowledge graph to perform causal reasoning on two types of event results to generate a reasoning enhancement result; and generating comprehensive abnormal analysis information and a response decision according to a reasoning enhancement result. According to the method, spatial-temporal features of multi-source data are fused, a causal reasoning mechanism is introduced, the identification and prediction results are subjected to correlation analysis, dynamic interpretation and response generation of complex events are achieved, and therefore the anomaly identification precision and the real-time performance and interpretability of risk decision making are improved.

Techniques for URL handling in Retrieval-Augmented Generation (RAG) systems are disclosed. A RAG system uses a generative AI response in the selection of links to include in a modified output from a RAG system. The links are based on URLs that are extracted from ingested documents. In response to a query, the system generates a response and then performs a matching operation to match a portion of the generated response to a URL in a URL-keyword mapping. The system finds a match between a portion of the generated response and a keyword that is associated with a URL, generates a hyperlink using the matched URL, and adds the hyperlink to the response.

The invention belongs to the technical field of intelligent interaction, particularly relates to an intelligent interaction method, system and device based on AI communication and a storage medium, and aims to solve the problems that in the prior art, AI scene coverage is narrow and is separated from a business process; the system comprises four modules: a context sensing and modeling module which collects multi-source context information in real time, and generates a three-dimensional context tensor through structured alignment and time sequence normalization; the intention analysis and reasoning module performs intention level decomposition by using a deep semantic model based on the tensor, and outputs an intention recognition result and probability distribution; the response generation and adaptation module is used for generating multi-mode responses such as texts, voices, images or equipment instructions in combination with intention and context tensors, and outputting the multi-mode responses after consistency verification; and the interactive feedback and optimization module collects user explicit and implicit feedback and is used for updating model parameters and strategy weights online to realize continuous optimization of the system.

The invention provides an automatic script generation system and method, computer equipment and a storage medium wherein an interface element pickup module is used for picking up target interface elements in batches and outputting element attribute information of each target interface element; the element type identification module is used for determining an element type label from a predefined element type set according to the element attribute information; the element screening module is used for carrying out screening, duplicate removal and semantic association processing on the target interface elements according to the element attribute information of the target interface elements and the element type labels to obtain a target element set containing element semantic attributes; and the script generation module is used for receiving a natural language instruction, analyzing and responding to the natural language instruction based on the target element set, and generating a target automatic script meeting a target RPA engine execution specification. By adopting the method, the automation degree, efficiency and accuracy of script generation can be improved.

The invention relates to the technical field of computers, and discloses an equipment protocol adaptive analysis and instruction scheduling method for an intelligent central control system, and the method comprises the steps: obtaining an original communication data flow of heterogeneous equipment, and extracting a protocol feature fingerprint to recognize a protocol type; calling a corresponding pluggable protocol parser to generate a standardized instruction object; analyzing instruction semantics and dynamically calculating priorities in combination with a system load; and performing non-blocking scheduling and execution monitoring on the instruction based on a resource token bucket mechanism. The system comprises a data acquisition module, a protocol identification module, a self-adaptive analysis module, a semantic analysis module, a priority calculation module, an instruction sorting module, a token management module, a scheduling execution module, an execution monitoring module and a response generation module. According to the method, through protocol self-learning, dynamic priority quantification and multi-dimensional resource isolation scheduling, the response certainty, expansibility and operation stability of the system in a high-concurrency scene are remarkably improved.

The invention relates to the technical field of semantic processing, and discloses a semantic processing system and method for home elevator intelligent customer service, and the system comprises a data collection module, a semantic processing module, a response generation and optimization module, and a communication and log module. The method comprises the steps of collecting natural language request information input by a user through voice, performing semantic analysis and intention recognition on preprocessed information based on a deep learning model, and synchronously recording an interaction log of a whole link according to an intention recognition result and the emergency degree and priority of optimized response content. In order to solve the problems that a traditional keyword matching system is poor in semantic comprehension ability and prone to misjudgment of user intentions, the context and real intentions of natural language requests can be deeply understood by introducing a semantic analysis model based on deep learning and combining a confidence degree evaluation mechanism, and when the system recognizes the low confidence degree condition, the user intentions can be accurately judged. And multiple rounds of clear conversations can be automatically triggered or interaction channels can be switched.

The present invention facilitates communication between sign language users and machines by translating sign language and text using AI models, deep learning computer vision, and word embeddings. Users interact via sign language, captured and processed through deep learning and NLP modules. The system converts sign language videos into text, constructs coherent sentences, and generates contextually appropriate responses using a Retrieve and Generate (RAG) model. Responses are translated back into sign language videos, spelling out words not found in the dictionary. If requested, a human agent can respond. Key features include high-accuracy recognition, context-aware response generation, dynamic vocabulary updates, and optional human interaction. The method ensures efficient processing with LLM, embedding techniques, and deep learning, optimizing translation accuracy and user experience. The system adapts to multiple languages and dialects by training on specific sign languages, making it applicable globally.

The invention discloses a vehicle information safety protection system based on combination of a national cryptographic algorithm and a PUF (Physical Unclonable Function), which belongs to the field of vehicle information safety and encrypted communication, and comprises a response generation module used for generating a PUF response in a safety chip of a vehicle; the key derivation module is used for generating an encrypted master key through a key derivation function based on the PUF response, the vehicle owner identity and the random number nonce; the encryption and signature module is used for generating a ciphertext and carrying out digital signature on the ciphertext; the state monitoring module is used for monitoring the running state of the vehicle hardware; the key management module is used for triggering a failure operation of the encrypted master key when the state monitoring module detects that the hardware state is abnormal; and the decryption verification module is used for regenerating the PUF response and verifying the consistency of the generated key so as to execute data decryption. According to the method, hardware-level encryption protection of the vehicle data is realized through combination of the PUF and the national cryptographic algorithm, so that the safety of the vehicle owner data is protected in the whole life cycle of the vehicle.

The invention discloses a network security situation awareness and automatic response decision-making system based on AI, and the system comprises a multi-source data processing module which is used for collecting multi-source data, executing time synchronization, field standardization and feature extraction, and generating a security event vector sequence; the situation modeling module is used for one-dimensional mapping and cross-source combination to form a situation representation vector; the attack relation modeling module is used for constructing a behavior combination structure and generating attack chain stage probability distribution and path contribution degree; the trend prediction module is used for extracting continuous time slices and inputting a time sequence modeling structure to generate a risk trend prediction result; the response strategy generation module is used for generating an optimal response action; and the closed-loop updating module is used for executing actions and updating parameters of each modeling module according to feedback information. According to the method, the multi-source security events are uniformly modeled based on the KAN network, so that collaborative updating of attack relation depiction, risk prediction and response generation is realized.

A system and method for improving generative artificial intelligence (AI) software application response is provided. The method includes: receiving a query directed to a generative AI software application; receiving a response to the query, the response generated by the generative AI software application; generating a first contextual value based on the received query; generating a second contextual value based on the received response; generating a verification score based on the first contextual value and the second contextual value; and initiating a mitigation action in response to detecting that the verification score is below a predetermined threshold

A dialogue system, comprising:an input, configured to receive input data from a user, wherein the input data comprises one or more of text data, speech data, image data and motion data; an output, configured to output data to the user; and one or more processors, configured to:obtain information identifying a skill and obtain information identifying a proficiency level of the user for the identified skill from stored proficiency level information; and execute at least one iteration of a coaching session, each iteration comprising performing one or more dialogue interactions, wherein each dialogue interaction comprises:receiving first input data from the user via the input;generating a first language model prompt and providing the first language model prompt to a language model, said first language model prompt comprising the first input data, the information identifying a skill, the information identifying a proficiency level of the user for the identified skill and a request to generate coaching information based on the first input data, the information identifying a skill and the information identifying a proficiency level; and generating first output data based on a first language model response to the first language model prompt and outputting, via the output, the first output data to the user;wherein the at least one iteration of the coaching session further comprises, after the one or more dialogue interactions:generating a second language model prompt and providing the second language model prompt to the language model, said second language model prompt comprising the information identifying a skill, the information identifying a proficiency level of the user for the identified skill, the first input data and the first output data, and a request to generate at least one proficiency update assessment based on the first input data, the first output data, the identified skill and the information identifying a proficiency level; generating second output data based on a second language model response to the second language model prompt and outputting, via the output, the second output data to the user; receiving second input data from the user via the input; determining a revised proficiency level of the user for the identified skill based on the second input data; andupdating the stored proficiency level information based on the revised proficiency level.

The invention discloses a honeypot automatic coping strategy generation method based on a large model. The method comprises the steps of S1, performing semantic analysis on dynamic attack behaviors; s2, performing context-aware threat reasoning; s3, adaptive strategy generation and semantic verification are carried out; s4, strategy executable compiling is carried out; s5, enhancing the efficiency of the closed-loop strategy; according to the method, the authority / service logic contradiction is thoroughly eliminated through a semantic consistency verification mechanism, so that the false alarm rate of the honeypot in the APT attack is reduced; an anti-recognition perturbation code injected by the low-entropy strategy compiling technology breaks through a traditional honeypot periodic response mode, and the fingerprint recognition success rate of an attacker is reduced; a resource penalty function of the Pareto optimal strategy sequence enables a trapping intensity mean value under limited resources to be improved; a double-channel updating mechanism promotes coevolution of a knowledge base and a constraint set, and the response generation speed for an unknown attack mode is shortened.

The invention relates to the technical field of electric power intelligent interaction, and discloses an AI large language model intelligent interaction system oriented to the electric power industry. The system comprises a multi-source data acquisition module for acquiring power industry multi-dimensional data from a plurality of heterogeneous data sources and integrating the power industry multi-dimensional data into a data warehouse; the semantic feature extraction module performs semantic analysis and feature extraction on the integrated data warehouse to obtain a semantic feature matrix; the industry knowledge graph construction module is used for constructing a power industry knowledge graph based on the semantic feature matrix; the user query analysis module is used for analyzing user query and extracting query feature vectors; the core semantic screening module is used for screening a core semantic sequence based on the semantic feature matrix and the query feature vector; and the intelligent response generation module generates an intelligent response based on the core semantic sequence and the power industry knowledge graph. The system can deeply understand user demands, generates professional and accurate responses, and is suitable for diversified intelligent interaction scenes in the power industry.

The invention provides a natural language interaction AI agent system for an intelligent tower crane and a method thereof. Wherein the agent system comprises a natural language understanding module, a task planning module, a tool scheduling engine, a large language model interface, a data adaptation layer, a result generation module and a learning optimization module; the proxy method comprises the following steps: S1, natural language input processing; s2, intention understanding and task planning; s3, data acquisition and preprocessing; s4, intelligent analysis and reasoning; s5, verifying and optimizing a result; s6, generating a natural language answer; and S7, feedback learning and optimization are carried out. According to the method, natural language interaction between the user and the intelligent tower crane system is achieved by constructing a special AI agent architecture, the large language model is automatically called to intelligently analyze the running state of the tower crane, and analysis results and decision suggestions which are easy to understand are provided for the user.

A query is received from a client device. A subset of documents relevant to the query is determined in part by determining an optimal configuration for the query. The subset of documents is inputted in a context window for a response generator and the query is inputted as a prompt for a query response. The query response received from the response generator is outputted to the client device.

The disclosed techniques automatically identify cyber-security attacks and predict attack next steps. Descriptions of previously observed cyber-attack campaigns are decomposed into attack campaign steps. Real-time security incident signals are generated by cybersecurity software. Attack campaigns are identified by mapping attack campaign steps to security incident signals. Custom-generated telemetry queries are executed to determine if a missing attack campaign step occurred. A machine learning model generates embeddings for attack campaign steps, security incident signals, and telemetry query responses. A security incident signal or a telemetry query response matches an attack campaign step when their embeddings are within a defined distance. A security alert may be raised when most or all of the attack campaign steps of a particular attack campaign are matched. Attack campaign steps that are not matched to security incident signals or telemetry query results are predicted as attack next steps.

Embodiments described herein relate to systems and methods for providing generative content for a graphical user interface of an issue tracking platform. The systems and methods can include generating a first prompt including a set of similar issues to a particular issue, and a request to extract actions performed for each issue of the set of similar issues. In response to providing the first prompt to a generative output engine, receiving a first generative response produced. In response to receiving the first generative response, a second prompt can be generated and include issue data extracted from the particular issue, the summary of actions received as part of the first generative response, and predetermined prompt language associated with a recommendation. A recommendation panel can be displayed and include content generated using the second generative response.

The invention relates to the technical field of supply chain management, in particular to a supply chain prediction and dynamic adjustment system based on deep learning, which comprises a data acquisition module, a data preprocessing module, a prediction model construction module, a risk assessment module, a decision support module, an execution module, a simulation and optimization module and a monitoring and reporting module. The prediction model construction module constructs a demand prediction model by using a deep neural network comprising LSTM and Attention mechanisms; the risk assessment module innovatively introduces a topological data analysis method, and realizes accurate identification and quantification of risks through three sub-modules of topological feature extraction, risk propagation modeling and intelligent response generation; the decision support module generates an optimal inventory strategy and a supply chain adjustment scheme based on the prediction data and the risk assessment result; the execution module is integrated with an ERP, a WMS and a supplier system through an API, a purchase plan is automatically executed, and high-precision prediction and dynamic adjustment of a supply chain are achieved.

A method for updating a question-answer mechanism includes obtaining, by a data system implementing the question-response mechanism, an input query, in response to the input query: generating a token sequence associated with the input query by applying a beam search algorithm on generated bound knowledge graphs, applying the input query to a large language model (LLM) of the question-response mechanism to obtain first outputs, applying the generated token sequence to a graph neural network (GNN) to obtain second outputs, applying fusion layers on the first outputs and the second outputs to generate a response associated with the input query and generated token sequence, and performing a remediation using the generated response, wherein the remediation comprises updating the LLM and the GNN based on the generated response.

Systems and methods for generating and providing outputs in a multi-device system can include leveraging environment-based prompt generation and generative model response generation to provide dynamic response generation and display. The systems and methods can obtain input data associated with one or more computing devices within an environment, can obtain environment data descriptive of the plurality of computing devices within the environment, and can generate a prompt based on the input data and environment data. The prompt can be processed with a generative model to generate a model-generated output. The model-generated output can then be transmitted to a particular computing device of the plurality of computing devices.

The invention provides a unified context and multi-level memory cooperation system and method for an LLM agent, and belongs to the field of artificial intelligence. The system comprises a unified context module which is used for aggregating and structuring all input information required for organizing agent operation, and providing an integrated context view for an LLM agent core; the LLM agent core is used for executing task planning, tool calling, result evaluation and response generation based on the context provided by the unified context module; and the multi-level memory system is connected with the unified context module and the LLM agent core and is used for hierarchically storing, managing and feeding back experience and knowledge of the agents. According to the method, the problems that an existing LLM intelligent agent is limited in context, low in memory efficiency and lack of self-correction capacity are solved, and the intelligent level and reliability of complex task processing are remarkably improved.

The invention relates to the technical field of digital human voice interaction, and discloses a digital human live voice interaction system fused with emotion calculation. The system constructs an emotional response time window by acquiring a user voice input data stream, the starting point of the emotional response time window is the end moment of the user voice input data stream, and the end point is obtained by subtracting the necessary duration of voice response synthesis from the preset maximum response cut-off moment. The system obtains an emotional state vector of the user in real time and judges whether the emotional state vector reaches an emotional intensity threshold value or not; and if so, predicting the total generation duration of the digital human voice response. And when the residual duration of the emotional response time window is equal to the total generation duration, taking the window starting point as a voice response starting moment, and controlling the digital human to start voice response generation. According to the system, the voice response matched with the emotional state of the user is generated through refined time window management and emotional state perception, invalid response and interaction delay are reduced, and the naturalness of digital human live broadcast voice interaction and the user experience are improved.

The present disclosure provides a system for adaptive voicemail and interactive voice response (IVR) detection in outbound calls. The system includes a call initialization module configured to establish an outbound call connection, a speech processing module configured to convert incoming audio signals into text in real-time, a classification module configured to analyze the text and determine whether the call has reached a live recipient, a voicemail system, or an IVR menu, a decision-making module configured to determine an appropriate course of action based on the classification, and a response generation module configured to generate and deliver appropriate responses based on the determined course of action. The system enables efficient handling of outbound calls by accurately detecting and responding to different call scenarios.

The invention belongs to the technical field of power distribution network operation and maintenance, and relates to a distribution network topological graph real difference diagnosis method based on a graph neural network, which comprises the following steps: injecting an initial detection signal sequence into a power distribution network, collecting and integrating responses generated by a collaborative sensing terminal, generating a graph node original feature set, and executing cross verification. Generating an enhanced initial topological connection feature graph, inputting the initial topological connection feature graph into a preset graph neural network model, and generating a topological diagnosis probability graph; according to the topology diagnosis probability map, generating a secondary detection signal sequence, injecting the secondary detection signal sequence into the power distribution network, generating a focusing response data set in the abnormal area, performing iterative diagnosis on the abnormal area based on the focusing response data set, and generating a final topology difference positioning report; according to the invention, the problem that a modern power grid is difficult to support large-scale and normalized topology verification requirements under the operation and maintenance requirements of real-time performance, accuracy and economical efficiency is solved.