173 results about "Generative systems" patented technology

The invention relates to an automatic assembly system and method for multi-scale digital twin components, and the system comprises a semantic relationship construction module which is used for carrying out explicit definition on the structural features, functional attributes, spatial layout requirements and logic dependency relationships of the multi-scale components, and constructing semantic relationships among the three types of components; the semantic reasoning and constraint engine module is used for carrying out logical reasoning through the semantic relationship constructed by the semantic relationship construction module and judging whether the component combination meets the assembly constraint or not; the assembly generation and configuration module is used for generating an assembly topological structure and a connection sequence of a system shelf according to the component candidate set output by the semantic reasoning and constraint engine module; and the man-machine interaction and visualization module is used for supporting a feedback closed loop between the engineer and the system and providing a visual display interface. The problems that an assembly method depends on artificial experience and is difficult to support high-frequency and multi-scene production line reconstruction are solved, and the method has higher semantic interpretation capacity, automatic combination capacity and context adaptive capacity.

The invention discloses a dynamic document template generation system and method based on large model workflow, and relates to the field of artificial intelligence technology, natural language processing and document automatic generation. Comprising the steps of 1, performing demand analysis on document demand description input by a user by adopting a pre-training language model and extracting document types, structural elements and business rules, 2, constructing a template structure according to an analysis result, distinguishing fixed contents and variable fields, defining conditional logic and synchronously displaying a template effect, and 3, constructing a template structure according to the analysis result. The method comprises the steps of receiving a user adjustment instruction and carrying out template optimization, 3, automatically matching variable fields with business database data, establishing dynamic association and automatically filling field contents according to the association relationship, 4, carrying out compliance detection on generated template contents and formats, prompting potential problems and enabling a user to carry out correction according to prompts. 5, according to the template confirmed by the user, the template scheme is converted into a construction instruction, and a standard format document is output; and 6, operation behaviors of the user are recorded, optimization features are extracted, the template library is updated, logic is generated, and continuous evolution of the template is achieved.

A method, computer program product, and computing system for processing a plurality of query-answer pairs associated with a generative artificial intelligence (AI) model. A first set of query variations are generated from the plurality of query-answer pairs using a genetic algorithm. A plurality of content portions associated with the first set of query variations are identified using a Retrieval Augmentation Generation (RAG) system. A fitness score associated with each of the query variations of the first set of query variations is determined using the plurality of content portions. A plurality of query variation-answer pairs are generated by generating a second set of query variations from the first set of query variations using the genetic algorithm and the fitness scores associated with each of the first set of query variations.

The invention discloses a domain knowledge-oriented interpretable and interactive question answering system and method, and the system comprises a user interface module which is used for receiving a natural language question inputted by a user, and outputting an answer and interaction information to the user; the retrieval enhancement generation core processing module is used for performing vector retrieval and map enhancement from a knowledge base based on a user question, generating context information for answer generation and submitting the context information to a large language model to generate an initial answer; the interpretability and interaction enhancement module is connected with the retrieval enhancement generation core processing module and is used for acquiring a processing log of the retrieval enhancement generation core processing module in real time and generating a visual reasoning path, and the interpretability and interaction enhancement module is further connected with the user interface module and is used for carrying out reverse clarification interaction and hypothesis reasoning. And performing closed-loop optimization and dynamic response of the driving system. By enhancing a traditional retrieval enhancement generation system, intelligent questions and answers with explainable process, clarifiable interaction and explorable analysis are realized.

Certain aspects of the disclosure provide a method for updating a retrieval augmented generation (RAG) system. The method includes receiving a user query and retrieving a set of datasets from an external knowledge base associated with a language model. The user query and retrieved datasets are provided to the language model as input tokens, which generates a response comprising output tokens. The system then extracts cross-attention weights from the language model, indicating how much attention each output token paid to each input token. Using these weights, the system generates attention scores for each dataset and identifies a top-k set of most attended datasets. If the generated response is determined to be relevant to the user query, the top-k most attended datasets are labeled as positive examples. The system then updates its parameters to prioritize retrieving these positive examples for future queries, enabling continuous self-supervised improvement.

The invention discloses a spray combustion sub-model integrated development method and equipment based on a large language model, and belongs to the technical field of engines. The method is realized by relying on a retrieval enhancement generation system and a multi-agent collaborative framework, and comprises the following steps: S1, user instruction and thesis extraction: extracting core information from related technical literatures to form a structured file according to a sub-module extension requirement proposed by a user; s2, source code analysis and comparison data generation; s3, function and interface planning: performing adaptive planning of an original model-OpenFOAM platform; s4, code generation and packaging: code conversion, dependency processing, test example generation and compilation packaging; s5, layered testing and verification graph generation; and S6, debugging and repairing: executing a positioning-repairing-regression closed-loop process through a debugging agent. According to the method, the problems that the spray combustion sub-model is difficult to migrate to the OpenFOAM simulation platform under different frameworks and the development period is long can be solved.

The invention relates to the technical field of natural language processing, in particular to a verifiable large model retrieval enhancement generation system and method based on an evidence chain, and the method comprises the steps: receiving an initial query, recognizing the fuzziness and information gap of the initial query in combination with an associated retrieval document, and generating a supplementary query set; based on the initial query, the supplementary query and the corresponding retrieval document, generating candidate answers with references and verifying the information supportability of the candidate answers; for the candidate answers passing the verification, extracting support information and constructing a hierarchical attribution mapping relation; and integrating the information to form to-be-evaluated information, and if the current verified to-be-evaluated information meets a preset sufficiency condition, integrating the generated preliminary answer and the to-be-evaluated information to synthesize a target answer. The method effectively overcomes the defects that a traditional RAG system is fragmented in information integration, has one-sided fuzzy query and answer, is low in attribution efficiency and excessively depends on retrieval content, and has the advantages of answer comprehensiveness, verifiability and deployment lightweighting.

Systems and methods are disclosed comprising instructions to collect vulnerability information over a network from a publishing source, collect network asset information of a communications network, generate a self-executing scanner agent configured to automatically scan the communications network for a known vulnerability based on an input including the collected vulnerability information and the collected network asset information, deploy the scanner agent at any network assets of the communications network that match a particular type of network asset indicated in the collected vulnerability information, generate a record including an indication of a particular network asset of the communications network in association with the known vulnerability in response to the scanner agent executing and detecting the known vulnerability in the particular network asset, and store the record in a data repository that aggregates records of detected known vulnerabilities in association with network assets of the communications network.

The invention discloses an FPGA test case automatic generation system and method based on a large language model. The system comprises a demand input module, a demand structuring module, a large language model case generation module, a case confirmation module, a verification execution module and a closed-loop optimization module. The large language model can automatically analyze the unstructured demand document and extract key information such as function points, input and output, boundary conditions and the like, and manual one-by-one interpretation is not needed. A big language model is guided through Prompt Engineering to dismantle complex requirements into test key points which cannot be subdivided, and it is ensured that all logic branches are covered by tests. The large language model can dynamically generate diversified test cases based on test key points, including a normal process, an abnormal process and an edge scene. And in combination with FPGA design specifications, historical error cases and other plug-in knowledge bases, the large language model can generate test cases closer to actual hardware behaviors.

The invention discloses a scenario generation method and system based on a large language model, and relates to the technical field of computers. By designing the scenario generation system comprising a scenario generation subsystem, a scenario setting subsystem, a scenario deduction subsystem and a scenario management subsystem, based on a large language model, a process of quickly generating a simulation scenario in a manner of retrieval enhancement generation or combat scenario text conversion is realized; moreover, the system has the functions of scenario editing, scenario deduction, scenario management and the like, achieves a workflow closed loop of intelligent generation, editing management, deduction inspection and modular storage, and improves the scenario generation efficiency and the intelligent level.

The invention provides a method for generating a system architecture diagram by automatically optimizing a cue word-driven reasoning large model. The method comprises the following steps of: preparing a multi-mode and reasoning large model and a sample data picture; generating content description and an initial layout requirement text; inputting the combined cue words into a large reasoning model, generating an HTML5 code, and converting the HTML5 code into a target picture; comparing the similarity between the target picture and the sample layout by the multi-modal model, and outputting a difference if the similarity does not reach a threshold value; and optimizing the layout requirement text based on the difference, and iterating until the similarity reaches the standard. Through cross-modal model collaboration, cue word closed-loop optimization, multi-role expert simulation and other technical innovations, the core pain points of strong manual dependence, insufficient flexibility and limited accuracy in traditional system architecture diagram generation are effectively solved, the technical spanning from passive auxiliary drawing to active intelligent generation is realized, and the system architecture diagram generation efficiency is improved. And the method has remarkable advantages in the aspects of automation degree, adaptability and generation quality.

The disclosed Retrieval Augmented Generation systems and methods include a system with several components. First, a user interface generates a query for a large language model (LLM). The system features prompt generator circuitry that accesses a database containing documents, each with priorities linked to various factors. This circuitry retrieves context and factor priorities from these documents in response to the query. The system also includes an LLM interface that submits a query to the LLM, incorporating the original query, retrieved context, and factor priorities. The system then receives a response from the LLM, which includes data related to the documents and factor priorities. Finally, an output interface presents the user with response data from the LLM, detailing information about the documents and the retrieved factor priorities.

The invention provides a personalized interactive intelligent report method based on a large language model. The method comprises the following eight steps: multi-modal prompt engineering, SQL expert model fine tuning, output correction and safety performance correction, intention perception and context construction, dynamic query generation and execution, multi-granularity insight extraction and narrative arrangement, interactive report presentation and multi-round dialogue. And a set of intelligent and adaptive data analysis report generation system is constructed. The system supports grassroots policemen to perform real-time interaction in a natural language form, automatically generates a structured query language (SQL) query or calls a preset application programming interface (API), and ensures query accuracy, security and high efficiency in combination with a multi-agent review mechanism and a dynamic knowledge graph. Meanwhile, the system has the deep data insight ability, narrative text abstracts and visual charts can be generated, and multi-round dialogue analysis is supported. The intelligent level and application efficiency of data analysis in the public security industry are remarkably improved, and the data use threshold is lowered.

A computer-implemented method in a hardware-secured mission-critical system governs text outputs produced by a transformer-based language model. A first transformer-based model generates candidate text and machine-identified propositions. A timing subsystem synchronized to PTP, GPS, or IMU clocks timestamps propositions and retrieved reports, logs, and sensor products, and admits only evidence within a configured skew window. Each proposition is segmented into semantic units under a constrained domain grammar that distinguishes subjects, predicates, modifiers, and domain-specific clauses. A second, architecturally isolated transformer-based model receives adversarial prompts built from propositions and summaries of admitted evidence and outputs structured challenge records with challenge types and severities. From evidence associations and challenge records the system constructs a semantic support mask and generates a conservative rewrite that preserves, in order, only tokens mapped to units that satisfy governance thresholds, introduces no new tokens, and passes a grammar-based integrity check before release as governed text output.

Techniques for evaluating Retrieval-Augmented Generation (RAG) systems are disclosed. A system performs a series of analysis operations associated with elements of a RAG system to evaluate the effectiveness of separate elements of the RAG system, and to evaluate the overall effectiveness of the RAG system. The system employs large language models (LLMs) and other analysis tools to generate metrics that indicate the effectiveness of the RAG system at various stages of operation. Based on these metrics, the system changes settings on the RAG system to improve performance.

The invention relates to the technical field of artificial intelligence, multi-agent systems and natural language processing, in particular to a multi-agent collaborative decision-making and communication system with a body based on a large language model, and the system comprises a sensing module which is used for obtaining environment and self state information; the large language model core processing module is integrated to each agent, comprises a multi-modal perceptual representation and semantic mapping system, a semantic-structured information bidirectional conversion network and a self-adaptive multi-path decision generation system, and is used for generating individual action decision suggestions and natural language communication contents oriented to other agents; the cooperative communication network is used for supporting efficient and semantic-rich interaction between intelligent agents based on natural language communication content; the environment coordination module is used for analyzing communication content and generating a global or local coordination instruction, a large language model is integrated into the multi-agent system, efficient communication and collaborative decision-making between agents based on natural languages are achieved, and the adaptability and decision-making quality of the system in a complex and dynamic environment are improved.

The invention relates to the technical field of intelligent document generation, and discloses an intelligent PPT content generation system based on semantic analysis. The system comprises a semantic analysis module, a content planning module, a template matching module, a content generation module and an adjustment module. The semantic analysis module processes a text input by a user and extracts key semantic elements; the content planning module is used for generating a multi-level outline, processing logic conflicts and recombining sub-topics; the template matching module extracts style parameters, compares template features and verifies the matching degree; the content generation module determines an optimal template and generates an initial PPT; and the adjusting module dynamically adjusts according to a user instruction to generate final content. According to the system, the content structure is optimized through semantic analysis, the template is accurately matched, the manufacturing process is simplified, the problems that in traditional PPT manufacturing, content organization is disordered, template selection is blind, modification and coordination are difficult and the like are solved, and the PPT manufacturing efficiency and quality are improved.

The invention discloses a power system-oriented dynamic knowledge base driven dialogue generation system, method, equipment and medium, and the system comprises a dialect collection and recognition module which is used for collecting dialect voice data under different regional power scenes, carrying out the noise reduction of the dialect voice data, carrying out the dialect recognition through a deep learning model, and obtaining a dialect recognition result; converting the dialect voice data into a standard text; a dynamic knowledge base module; the language processing and image reasoning module is used for receiving the standard text, performing semantic understanding, generating semantic representation in combination with a knowledge base in the dynamic knowledge base module, and performing target detection and recognition on a power equipment fault related image input by a user to obtain an image recognition result; and a dialogue generation module. According to the invention, a cooperative system of four modules of dialect acquisition and identification, a dynamic knowledge base, language processing and image reasoning and dialogue generation is constructed, so that intelligent dialogue service oriented to the power industry is realized.

A method, computer program product, and computing system for processing a plurality of query-answer pairs associated with a generative artificial intelligence (AI) model. A first set of query variations are generated from the plurality of query-answer pairs using a genetic algorithm. A plurality of content portions associated with the first set of query variations are identified using a Retrieval Augmentation Generation (RAG) system. A fitness score associated with each of the query variations of the first set of query variations is determined using the plurality of content portions. A plurality of query variation-answer pairs are generated by generating a second set of query variations from the first set of query variations using the genetic algorithm and the fitness scores associated with each of the first set of query variations.

Disclosed herein are a dialogue generation system that reflects the location of a character within a virtual space, and a dialogue generation system that provides a more immersive and richer dialogue experience to a user by providing metadata including location information of objects within a virtual space to an API server together with a user's query at the time the user's query is received, thereby enabling a generative AI model to generate more natural and contextual dialogue by considering the character's location.

The embodiment of the invention provides a reply information generation method and system, a reply information generation model training method, electronic equipment, a storage medium and a computer program product. The method comprises the following steps: executing a first task based on context information of target information and user information associated with the target information to identify a user intention matched with the target information, and extracting key semantic slot position information associated with the target information; executing a second task based on the user intention and the key semantic slot position information to obtain a reply strategy for the target information; and executing a third task based on one or more of the target information, the context information, the reply strategy and the key semantic slot position information to generate reply information for the target information. According to the method, the generation complexity of the reply information is reduced, the generated reply information has better action guidance, and the quality of the generated reply information is improved.

The invention relates to a question and evaluation generation system and method fusing expert preference meta-cognitive intelligent agents, and belongs to the technical field of intelligent education. The system comprises an automatic question generation module used for calling a large language model to generate questions and evaluations according to a subject information base and an expert preference model; the meta-cognitive agent evaluation and optimization module is used for calling a meta-cognitive agent to perform multi-dimensional evaluation on questions and evaluation according to the expert preference model and generating an evaluation result and an optimization suggestion; and the expert interaction module is used for obtaining feedback information of the expert on the questions, the evaluation, the evaluation results and the optimization suggestions, and if the feedback information shows that the expert does not pass the verification, adjusting model parameters of the expert preference model, the large language model and the meta-cognitive agent according to the feedback information until the expert passes the verification. According to the invention, a deep cooperation mechanism between experts and artificial intelligence is established, artificial intelligence is guided to learn expert experience and preference through interaction, and high-quality questions and evaluation and assessment results are generated.

The invention discloses a compliance report generation system for automatically analyzing financial regulation texts into executable rules, and belongs to the technical field of artificial intelligence and financial science and technology. The system comprises a regulation text obtaining and preprocessing module used for obtaining at least one structured clause text; the atomization rule unit analysis module is used for analyzing the structured term text into an atomization rule unit; the knowledge graph and logic context tree construction module is used for constructing a regulation knowledge graph and a logic context tree; and the executable rule synthesis module is configured to retrieve a group of target atomization rule units and synthesize at least one executable rule expression based on the group of target atomization rule units. According to the method, by introducing the atomization rule unit and the logic context tree, the accuracy and query efficiency of law and regulation analysis are improved, and automatic and high-precision logic conversion of complex financial laws and regulations is achieved.

The invention discloses a computer algorithm visualization teaching generation system based on a large model, and relates to the technical field of artificial intelligence-driven education. The work content of the system comprises the following steps: constructing a mapping relationship between algorithm categories and features, generating an understandable feature set, and labeling difficulty levels; guiding a user to select an algorithm category and program a basic label through a graphical algorithm classification graph; k rounds of dynamic questions and answers are adopted, questioning features are screened based on an information entropy gain ratio, and answer vectors are constructed in combination with user answers; through cosine similarity matching algorithm feature vectors, teaching content is recommended, difficulty is adjusted, or non-questioning features are supplemented and inquired; and converting the user data into a natural language text, calling a preset large model interface to generate a learning suggestion, and performing interactive visual display. According to the system, a whole-process closed loop of algorithm teaching from data acquisition and preprocessing to dynamic generation and personalized recommendation is realized, and the teaching precision and efficiency are improved.

The present invention makes it possible to create desired scenario data in a simulation of autonomous driving, for example, with reduced labor.　This scenario data generation system includes a storage device and a processor. The storage device stores scenario data that includes information about operations of one or more objects at each sample time in a scene in which the one or more objects operate, and prompt data that instructs a generative artificial intelligence to perform a desired operation of each object. The processor starts a step-by-step simulation at each sample time based on the scenario data, causes, in each step, the generative artificial intelligence to generate a next state that is a state of the object at the next sample time due to the desired operation based on the state of the object in the scenario data, updates the state of the object at the next sample time in the scenario data to the next state, and outputs the scenario data when the simulation is completed.

The application relates to the technical field of natural language processing, in particular to an evidence chain-based verifiable large model retrieval enhancement generation system and method, which comprises the following steps: receiving an initial query, identifying the ambiguity and information gap of the initial query in combination with associated retrieval documents, and generating a supplementary query set; generating a cited candidate answer based on the initial query, the supplementary query and the corresponding retrieval document, and verifying the information supportability of the cited candidate answer; extracting supportable information from the verified candidate answer and constructing a hierarchical attribution mapping relationship; integrating the above information to form to-be-evaluated information; if the to-be-evaluated information currently verified satisfies a preset sufficiency condition, integrating a generated preliminary answer and the to-be-evaluated information to synthesize a target answer. The application effectively solves the defects of the traditional RAG system, such as information integration fragmentation, one-sided query answering, inefficient attribution and excessive dependence on retrieval content, and has the advantages of comprehensive answer, verifiability and lightweight deployment.

The invention relates to a planetary gear box fault diagnosis algorithm automatic generation system and method based on a large language model, and the system comprises a large language model generation module which is used for receiving cue words, selecting a parent planetary gear box fault diagnosis algorithm and architecture component knowledge, and combining the characteristics of a planetary gear box fault diagnosis task, calling a large language model to generate a new planetary gearbox fault diagnosis algorithm code; the dynamic code execution and verification module is used for code execution and verification; the architecture evaluator module is used for performing training and performance evaluation; the intelligent evolutionary strategy module is used for analyzing the training history to obtain an analysis result; the program database module is used for storing and managing populations of planetary gearbox fault diagnosis algorithms and constructing cue words; and the evolutionary control engine is used for coordinating and controlling the working process of each module and controlling the number of iterative evolutionary times to realize evolutionary circulation. Compared with the prior art, the method has the advantages of automatic generation, iterative evolution and the like.

The invention discloses a defense method for a retrieval enhancement generation system, electronic equipment and a medium. The method comprises the following steps: responding to a problem, retrieving a plurality of documents output by an enhancement generation system, and compressing and aggregating all the documents; in response to the problem and the aggregated document, the large language model outputs a denoising principle for the retrieval enhancement generation system; when the de-noising principle accords with the consistency evaluation, constructing a basic principle learner, and performing rational generation learning on the basic principle learner according to the problem and the de-noising principle; in a reasoning stage, inputting the problem and the compressed and aggregated document into the trained basic principle learner, and outputting a denoising principle; the retrieval enhancement generation system generates explainable and accurate answers according to questions, de-noising principles.

The invention relates to the technical field of data enhancement and system engineering, in particular to a user behavior data enhancement method of a large language model based on RFLP driving. Comprising the steps of determining a core driving problem and an expected application scene of to-be-generated user behavior data, determining a task target, a data structure requirement, a data characteristic requirement, a constraint condition and a non-functional requirement of a large language model data generation system based on RFLP driving according to the core driving problem and the expected application scene, and encoding the requirement specifications into a plurality of functional modules; and forming a data generation function set according to the modules, constructing a logic scheme based on'strategy-component-combination 'for the system based on the function set, converting the logic scheme into an executable scheme, finally controlling the large language model to execute the executable scheme, and generating to-be-generated user behavior data. Therefore, the problem that the cue word is difficult to ensure the quality and consistency of the generated data is solved, and a scientific methodological support is provided for constructing a high-quality, controllable and extensible synthetic user data generation system.

The application discloses an intelligent imaging generation system and method, which realizes closed-loop control of spraying quality by real-time linkage adjustment of spraying mechanical arm and imaging equipment parameters through a dynamic parameter adjustment module, and combines a defect prediction and optimization decision module and a multi-modal process knowledge base. The system integrates a high-resolution visual sensor, a capacitive thickness detector and a temperature and humidity sensor, fuses multi-source data by using a Kalman filtering algorithm, and dynamically adjusts the speed of the mechanical arm, the nozzle pressure and the imaging gain coefficient. The defect prediction module simulates coating defect morphology based on a physically constrained generative adversarial network and generates a process optimization instruction set through reinforcement learning. The multi-modal process knowledge base stores an associated model of material characteristics, environmental data and defect patterns, supports incremental learning and historical data decay update. Through dynamic parameter collaborative optimization, defect prediction prepositioning and multi-modal data fusion, the spraying quality detection efficiency and coating consistency are improved, and the rework rate is reduced.