258 results about "Generative systems" patented technology

The invention provides an agent-based large language model retrieval enhancement generation system and method, and the system comprises a planning layer which is used for receiving user query, carrying out the multi-round iterative decomposition of a complex task through a task planning agent, and generating an atomic query or a direct response; the execution layer is used for executing the atomic query generated by the planning layer in parallel, calling a search module to obtain external knowledge base data, and caching an intermediate result through a memory module; the answer detection module is used for performing multi-dimensional detection on the generated result, including preference, accuracy, integrity and logicality; the dynamic decision-making module is used for adaptively adjusting a subsequent retrieval strategy and a task planning process according to a detection result and user feedback; and a cross-layer interaction mechanism enables the planning layer and the execution layer to realize collaborative optimization through context sharing and iterative feedback.

Embodiments of the present invention provide an innovative Retrieval-Augmented Generation (RAG) system tailored for financial analysis, significantly enhancing the precision and contextual relevance of Large Language Models (LLMs). A part of the system is a query augmentation component that leverages a knowledge graph to semantically enrich user queries, ensuring comprehensive retrieval of pertinent financial documents. A noise filtering mechanism refines the search results, while a relevance ranking component prioritizes documents based on context (e.g., user and task). The system employs prompt engineering to guide the LLM in generating responses that meet the specific requirements of financial analysis. Additionally, the LLM is fine-tuned using a corpus of financial questions and answers, reinforced by human-in-the-loop feedback, to adapt the model to the financial domain's unique linguistic and structural nuances. This advanced RAG system offers financial professionals timely, reliable, and actionable insights, providing a competitive edge in a rapidly evolving financial landscape.

An unstructured data query-response pair generation system (generation system) populates a knowledge base of query-response pairs for queries of natural language content in unstructured data by prompting a first large language model (LLM) text extracted from the unstructured data. An unstructured data chatbot (chatbot) leverages the knowledge base by augmenting prompts to a second LLM responding to user queries for natural language content in the unstructured data with query-response pairs having queries that are semantically similar to the user queries. The knowledge base and LLMs are updated based on user feedback correcting responses, continually improving quality of the generation system and chatbot.

The invention discloses a multi-modal natural language understanding and generating system and method. The method comprises the following steps: constructing a cross-modal pre-training module, training a multi-modal encoder, and establishing a cross-modal association mapping space; mixing prompt fine tuning is carried out, and a complete blank filling template is constructed; according to the intention reasoning network, extracting multi-round dialogue intention representation of the user, and retrieving an external knowledge base for fine-grained reasoning; constructing a unified semantic representation framework, embedding the text, the image and the voice into a unified space, and generating a query vector of multi-modal intention perception; and the knowledge query module based on key value memory generates entity-level multi-modal replies and optimizes the semantic comprehension and generation capability of the dialogue model. According to the method, the multi-modal information understanding and generating capacity is improved, deep association and understanding of image and text information are achieved, downstream task adaptability is enhanced, task completion accuracy and efficiency are improved, unified semantic representation of the multi-modal information is achieved, and support is provided for information retrieval and utilization.

The invention discloses a SysML model automatic generation system based on mixed AI and domain knowledge, and the system comprises a preprocessing module which is used for carrying out the text preprocessing and structural enhancement of an engineering document of a PDF or Word version; the NLP extraction module is used for identifying six types of core entities by adopting aviation corpus fine tuning BERT, constructing a document-level relational graph by utilizing GNN, modeling a cross-paragraph dependency relationship, calling LLM for semantic fuzzy sentences to generate a thinking chain, extracting a reasoning path and solving ambiguity; the rule conversion engine module is used for mapping the entity relation graph into a SysML memory object tree; and the controllable generation module is used for carrying out limited decoding on the LLM by utilizing a Guidance framework. The invention further discloses an automatic SysML model generation method based on the mixed AI and domain knowledge. According to the method, the problems of low manual modeling efficiency and poor semantic consistency in traditional MBSE implementation are solved.

The invention discloses a report generation system and method based on a legal knowledge graph, and relates to the technical field of artificial intelligence and law, and the method comprises the steps: obtaining a legal appeal of a user through a non-standardized domain corpus collection engine, carrying out the multi-dimensional intention deconstruction through a semantic slicing model, and obtaining a judicial element topology network; inputting the judicial element topology network into a law deduction engine of dynamic value perception ability, calculating a decision vector weight through a dynamic value evaluation network, and generating a judicial decision path by using a time sequence feedback mechanism; based on a judicial decision path, through an association rule mining engine of a multi-modal knowledge graph, a dynamic weighting law retrieval confidence matrix is generated in combination with semantic coupling degree analysis and logic path influence evaluation. According to the method, qualitative change of legal consultation from information retrieval to intelligent decision is realized through collaborative innovation of a legal deduction engine with dynamic value perception capability and a legal semantic constraint decoder.

The invention relates to the technical field of intelligent generation, and particularly discloses a decoration scheme generation system and method based on generative artificial intelligence, which utilize an artificial intelligence technology based on deep learning to analyze the demand of a user, extract the semantic feature of the decoration demand of the user, and meanwhile, generate the decoration scheme according to the semantic feature of the decoration demand. And feature learning is performed on a large number of indoor decoration elements, and an indoor decoration element feature library is established, so that a room decoration effect picture is intelligently generated based on the semantic association interaction features of the indoor decoration element set and the decoration demand of the user. Therefore, the room decoration effect picture can be automatically and intelligently generated for the user to refer to and select according to the personalized requirements of the user, so that the efficiency and the quality of decoration design are improved, and better decoration experience is provided for the user.

The invention provides a legacy system-oriented large-model-driven API document automatic generation system, belongs to the crossing field of artificial intelligence and software development, and provides a multi-modal data fusion and closed-loop verification mechanism aiming at the defects of a traditional API document generation method in the aspects of semantic comprehension, dynamic context capture and multi-technology stack adaptation. A code static feature and a dynamic track during operation are analyzed through a multi-source data acquisition module, and an interface semantic feature is extracted in combination with a field self-adaptive large model of a semantic enhancement analysis module; deducing an implicit service rule by fusing static / dynamic characteristics through a graph neural network, and generating a standardized document conforming to an OpenAPI specification through a parameterized template generative adversarial network (PT-GAN); and finally, performing three-level verification and closed-loop optimization through a sandbox environment. The method supports a heterogeneous system of 16 programming languages such as Java / C + + / Python, interface version changes can be automatically recognized, document patches are generated, the problems of missing and outdated system documents and low maintenance efficiency are solved, and maintainability and integration efficiency of enterprise-level systems are remarkably improved.

The invention discloses an intelligent crawler generation method and system based on a large language model and an MCP protocol, belongs to the technical field of network data collection, and solves the problem that the capability of LLM in dynamic webpage analysis and anti-crawling strategy generation links cannot be fully exerted due to the fact that LLM and browser interaction protocols cannot be effectively integrated in the prior art. The method comprises the steps of analyzing an acquisition demand based on a large language model and generating a standardized demand description document, realizing interaction between the large language model and a browser based on an MCP protocol, analyzing a page complete DOM tree structure through a crawler script generation system, and performing quality verification and intelligent repair on a generated crawler script. According to the method, the complete DOM tree and the dynamic data rendered by the browser are obtained through the MCP, and the large language model can be called to automatically analyze the element positioning strategy, so that the collection script is adaptively generated, and high efficiency, intelligence and automation of webpage data collection are ensured.

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 provides an industrial defect image generation system and method based on deep learning, and relates to the technical field of artificial intelligence. A defect form adaptive module, a physical attribute modulation module and a multi-mechanism fusion module are integrated in a defect image generation module; dynamically selecting a feature extraction unit according to the defect type label based on a pre-constructed generative network model, and generating a defect feature map; generating an affine transformation parameter based on the physical attribute vector, and performing channel-by-channel linear modulation on a defect feature map of a middle layer of the generative network model, so that a multi-scale defect feature map finally generated by the generative network model contains specified defect type features and physical attribute features; and fusing the multi-scale defect feature image and the defect-free background image to obtain a fused defect image. The problems that in the prior art, industrial defect image generation is insufficient in sense of reality, poor in controllability and poor in fusion effect are solved, the method can be used for data enhancement of industrial visual inspection, and downstream model performance is improved.

According to the intelligent imaging generation system and method, parameters of a spraying mechanical arm and imaging equipment are adjusted in a linkage mode in real time through a dynamic parameter adjusting module, and closed-loop control over the spraying quality is achieved in combination with a defect prediction and optimization decision module and a multi-modal process knowledge base. A high-resolution visual sensor, a capacitive thickness detector and a temperature and humidity sensor are integrated in the system, multi-source data are fused through a Kalman filtering algorithm, and the mechanical arm speed, the nozzle pressure and the imaging gain coefficient are dynamically adjusted. The defect prediction module simulates a coating defect form based on a physical constraint generative adversarial network and generates a process optimization instruction set through reinforcement learning. The multi-modal process knowledge base stores a correlation model of material characteristics, environmental data and defect modes, and supports incremental learning and historical data attenuation updating. And through dynamic parameter collaborative optimization, defect prediction preposition and multi-modal data fusion, the spraying quality detection efficiency and the coating consistency are improved, and the rework rate is reduced.

The invention relates to the technical field of intelligent abstract generation, in particular to an intelligent archive abstract generation system and method based on a natural language processing technology. The system comprises a multi-layer semantic generation core unit which performs semantic hierarchical segmentation analysis on an original file text to construct a multi-layer semantic graph, and constructs an abstract generation model to generate abstract content; the knowledge graph fusion engine unit associates and extracts terms in an original file text, constructs an entity mapping relation, and sets semantic graph node expression weights in an abstract generation model according to semantic association confidence scores; a paging index cache retrieval unit performs fragment division processing on each node in the multi-layer semantic graph, and constructs an abstract content fragment index structure containing a node set; and the version tracing transaction management unit carries out version recording on generation and modification operations of the abstract contents. The invention discloses an intelligent archive abstract generation system which is constructed by fusing a multi-layer semantic graph and is associated with a knowledge graph.

A method, computer program product, and computing system for processing a query provided to a generative AI model. A content portion retrieved by a Retrieval Augmented Generation system for the query is processed. User context information associated with a user providing the query is determined. Hyperparameters are generated for processing the prompt with the generative AI model by processing the query, the content portion, and the user context information using run-time surrogate model inversion optimization.

The invention relates to the technical field of artificial intelligence and software development, in particular to a DevOps assembly line intelligent generation system based on natural language description. According to the intelligent generation system for the DevOps assembly line based on the natural language description, an interaction layer is responsible for visually editing an interaction interface and outputting a generated executable script and a flow chart to a user; the intelligent processing layer takes Python and Go as development languages, fuses a knowledge graph technology, supports a dynamic code generation capability and realizes intelligent business process management and automatic execution; the infrastructure layer comprises three core modules, namely a metadatabase, a sandbox environment and an execution engine, and is responsible for providing basic support and resource guarantee for system operation. According to the intelligent generation system for the DevOps assembly line based on the natural language description, the configuration threshold of the DevOps assembly line is remarkably reduced, the cross-platform compatibility is ensured, the safety and reliability of the assembly line are greatly improved, and production accidents caused by manual configuration errors can be effectively avoided.

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.

The invention discloses a multi-language advertisement creative generation system, and the system comprises a data collection and preprocessing module which is responsible for collecting text travel and language data, and carrying out the cleaning and marking. The culture feature analysis and modeling module extracts culture features in the data and constructs a culture model to map features and language expressions. The creativity generation module constructs a semantic understanding model based on deep learning, is in butt joint with the culture model, and generates creativity themes and plots. And the language generation module inputs the creative theme and the target language into a language generation model and outputs an advertisement copywriting. And the multi-language optimization and integration module optimizes the copywriting of each language and integrates the copywriting into an advertisement creative set. And the user interaction and feedback module provides an interface, inputs text travel items, target languages and parameters, outputs copywriting, and optimizes the system according to user feedback. According to the invention, the generation efficiency of the advertisement creativity can be improved, the adaptability of the advertisement copywriting in different language culture environments is enhanced, and the limitation of traditional advertisement creativity generation can be broken through.

The invention discloses a large model query generation system and method combining graph structure analysis and execution, and the system comprises a graph mode standardization module which converts a description file into a unified structured representation; the large language model module is used for screening out nodes, edges and sub-graph modes required by query statements according to the natural language problem and the structured representation of the user; the graph mode detection module is used for detecting and modifying the screened nodes, edges and sub-graph modes through scripts; the large language model module generates an initial query statement according to a natural language problem of a user and detected and modified nodes, edges and subgraph modes; the query statement detection module is used for detecting the initial query statement through a script and generating an error report; and the large language model module generates a modified query statement. According to the large model query generation system and method combining graph structure analysis and execution, the accuracy of graph mode selection is remarkably improved through the thinking ability of the large language model and the script error detection ability, and therefore the quality of graph database query statement generation is improved.

Disclosed are devices, systems, and techniques for evaluation of machine learning models, pipelines of machine learning models, retrieval-augmented generation (RAG) systems, and / or other artificial intelligence systems. Example techniques include receiving, from a client device, an evaluation task to evaluate a language model (LM) using a plurality of evaluation benchmarks (EBs) associated with respective EB dataset and configuring, using an evaluation API, respective sets of evaluation jobs to implement the evaluation task. An individual set of evaluation jobs is configured to evaluate, using the corresponding EB dataset, performance of the LM to obtain a set of evaluation metrics. The techniques further include executing the sets of evaluation jobs to obtain respective sets of evaluation metrics and causing, using the evaluation API, a representation of the sets of evaluation metrics to be provided to the client device.

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 relates to the technical field of advertisement generation, in particular to a cross-border e-commerce personalized advertisement generation system based on a generative adversarial network. The method comprises the following steps: S1, acquiring first information input by a user in real time through an acquisition module; s2, outputting a first result based on the first information through a multi-modal generation module; s3, evaluating the first result for at least one time through a multi-dimensional judgment module so as to optimize the first result and generate a second result; s4, dynamically optimizing the second result through a dynamic optimization module, and generating a third result; according to the cross-border e-commerce personalized advertisement generation system based on the generative adversarial network, cross-border e-commerce personalized advertisements are generated, so that the investigation, communication, planning and design time of users can be shortened, the generation efficiency is greatly improved, and the manufacturing cost is greatly reduced.

The invention relates to an AI-based virtual character generation method and device, a terminal and a storage medium, and the method comprises the steps: obtaining a corresponding multi-modal feature in response to a character generation instruction input by a user side; inputting the multi-modal features into a pre-trained multi-modal fusion model, and enabling the multi-modal fusion model to output and generate role features; performing semantic enhancement and logical reasoning on the generated role features through a knowledge graph to obtain enhanced role features; generating initial virtual character data based on the enhanced character features and sending the initial virtual character data to the user side for adjustment; and when optimized virtual character data fed back by the user side is received, generating a corresponding virtual character. According to the virtual character generation system and method, by combining the multi-modal feature extraction, knowledge graph reasoning and generative AI technologies, the bottleneck of an existing virtual character generation system on the intelligent level is effectively solved, and the effect of improving the virtual character generation quality and the user experience is achieved.

The present disclosure describes a generative artificial intelligence-based solution for generating a narrative associated with a transaction. The generative artificial intelligence described herein acquires one or more data points associated with the transaction. Based on these data points, the generative artificial intelligence may generate one or more narratives for the transaction. The one or more narratives may be provided to a user device for a user's review and / or approval. After a narrative is approved, the narrative may be stored. If the transaction is later contested, the narrative may be provided to user to refresh their recollection about the circumstances regarding the transaction.

The invention discloses a modular content generation method and system for embedded teaching, and relates to the technical field of teaching semantic generation. The invention discloses a modularized content generation system for embedded teaching. The modularized content generation system comprises an embedded teaching module division module and an embedded teaching content generation module, according to the method, the problems of loose organization and poor reusability of the traditional embedded teaching content are solved through semantic analysis and structured processing of the original embedded teaching content; by introducing a skill dependence analysis mechanism, a clear preposed skill set and a clear output skill set are distributed for each initial teaching module, a teaching path network oriented to capability growth is constructed, and the definition of teaching logic and the controllability of a learning path are improved; learning fatigue evaluation is carried out by fusing multi-dimensional factors such as skill span, path complexity and cognitive load, and cognitive overload or content redundancy in the learning process of students is effectively avoided.

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.

Systems and methods of adaptive context partitioning in a retrieval-augmented generation system, including receiving a query from a user; forming a combined context by retrieving relevant documents from a document database based on the query, determining characteristics of the combined context, monitoring current system resources including processor availability and memory utilization, dynamically determining a partition size based on the current system resources and the one or more characteristics of the combined context, partitioning the combined context into context partitions according to the partition size, generating intermediate analysis results by processing each context partition using a mapper prompt large language models (LLMs), generating a final response by processing the intermediate analysis results using a reducer prompt through the LLMs, and transmitting the final response to the user.

The invention relates to an access control strategy automatic generation system and method based on multi-task learning, and the method comprises the steps: carrying out the word segmentation, cleaning and embedded vector conversion of an original access control text through a data preprocessing module, and constructing a normative input format; the feature sharing layer module is used for extracting deep semantic features of a text through multi-layer bidirectional coding and an attention mechanism and providing unified representation for downstream tasks; the access control statement identification module is used for judging whether each sentence in the text is an access control statement or not and realizing automatic identification of strategy related contents; and the attribute extraction and annotation module is used for annotating words in the access control statements and extracting subject, object and operation access control attributes. A word coding layer and a sentence coding layer are shared, local and global attention mechanisms are combined, key information of a text is extracted, the semantic understanding ability is enhanced, and cooperative training of statement recognition and attribute extraction is achieved; a conditional random field CRF structure is used for sequence labeling, and the structural rationality of attribute labels is ensured.

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.