Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

51 results about "Private knowledge" patented technology

Cooperative scheduling method based on security agent

The invention discloses a collaborative scheduling method based on a security agent, and relates to the technical field of network security design. The specific operation of the security agent collaborative scheduling method comprises the steps of system deployment and initialization, security operation task execution process, agent self-learning and capability evolution implementation, cross-domain security agent collaborative adaptation implementation and security operation visualization and traceability implementation. Connection paths between the security agent and business data, a private knowledge base, a security tool and multiple models are broken through, the problem that all elements in a traditional mode lack efficient communication and collaboration is solved, and by means of agent routing dynamic scheduling, A2A protocol interaction, MCP protocol tool calling, RAG business interface calling and multi-model combination, the security of the security agent is improved. Integration and intellectualization of the safety operation process are achieved, the continuity and the response speed of the operation process are improved, safety operation is more efficient and collaborative, and complex and variable safety requirements are met.
Owner:SHANGHAI DIGITAL SECURITY TECH CO LTD

Private knowledge question-answering method in combination with knowledge graph

The invention discloses a private knowledge question-answering method combined with a knowledge graph. Comprising the following steps: S1, collecting private basic corpora; s2, fusing the acquired basic corpus information; s3, establishing a knowledge graph; s4, partitioning the vector data information; s5, constructing a knowledge graph; s6, processing the problem of the user; s7, user question answer retrieval and generation; s8, outputting answers to the user questions; the knowledge graph is constructed and combined with the large language model to deeply understand and efficiently retrieve complex information in the private knowledge base, and the knowledge graph organizes entities and relationships, so that the system can better capture association between the information, the understanding ability of a user to put forward a question is enhanced, and the user experience is improved. The combination not only improves the accuracy and efficiency of information acquisition, but also provides new assistance for innovation and development of enterprises.
Owner:WUXI URBAN BLOCKCHAIN ADVANCED RES CENT +1

Management method of multi-mode enterprise knowledge base system

The invention provides a management method of a multi-modal enterprise knowledge base system, which comprises the following steps: integrating multi-modal data in an enterprise, constructing a private knowledge base, and combining the private knowledge base with a public knowledge base; performing intention analysis on a multi-modal retrieval demand submitted by the user based on the joint result, performing condition retrieval on the private knowledge base and the public knowledge base based on an intention analysis result, and outputting multi-modal query data; and carrying out quality inspection on the condition retrieval process, carrying out dynamic optimization on a joint result based on a quality inspection result to obtain a final multi-modal enterprise knowledge base, and carrying out deployment management on the multi-modal enterprise knowledge base. The management effect of the multi-modal knowledge in the enterprise is greatly improved, and the retrieval accuracy of the knowledge is also improved.
Owner:SHENZHEN YUNZHIYIN TECH CO LTD

Intelligent teaching assisting system base LLM training method for well drilling simulator

The invention provides an intelligent teaching-assistant system base LLM training method for a drilling simulator, and belongs to the technical field of petroleum drilling large language model training, and the method comprises the steps: obtaining the simulation data of the drilling simulator based on an intelligent teaching-assistant system base large language model, and carrying out the field self-adaptive pre-training; performing fine adjustment on the original weight by utilizing low-rank self-adaption, and performing operation steps and fault diagnosis generation; the method comprises the following steps: constructing an enterprise private knowledge base, training a large language model by utilizing retrieval enhancement generation, and obtaining a trained verification header through explicit supervision; a direct preference optimization loss function is defined, a direct preference optimization loss function of retrieval perception is obtained in combination with the verification head, the large language model is optimized, and large language model training is completed; according to the invention, an external professional knowledge base can be fused, an equipment operation mechanism can be understood, an intelligent assistant architecture with teaching guidance and error diagnosis capabilities is provided, and the intelligent level of the drilling simulator is improved.
Owner:SOUTHWEST PETROLEUM UNIV +1

Code conversion method and device, storage medium and program product

The embodiment of the invention provides a code conversion method and device, a storage medium and a program product. The method comprises the following steps: segmenting a to-be-converted code file into a plurality of code snippets, retrieving a conversion rule matched with each code snippet from a preset conversion rule library, and retrieving a knowledge fragment matched with each code snippet from a user private knowledge base; based on the multiple code snippets, the conversion rules matched with the code snippets and the knowledge snippets matched with the code snippets, target cue words are generated, the code conversion model is guided through the target cue words, and the code snippets to be converted are recognized from the multiple code snippets; the to-be-converted code snippets are rewritten on the basis of the conversion rules matched with the to-be-converted code snippets and the knowledge snippets matched with the to-be-converted code snippets, so that the to-be-converted code file can be converted into the target code file under the target version more accurately by the code conversion model; and the accuracy of a code conversion result between different versions is improved.
Owner:BEIJING 58 INFORMATION TTECH CO LTD

Workflow generation method and device fusing large language model and knowledge base

The invention belongs to the technical field of artificial intelligence, provides a workflow generation method and device fusing a large language model and a knowledge base, and aims to solve the problem of low accuracy of automatic workflow generation based on the large language model in the traditional technology. Identifying multi-dimensional context information contained in the multi-dimensional context information; retrieving an enterprise private knowledge base according to the multi-dimensional context information to obtain domain knowledge fragments related to workflow generation; the workflow requirement description, the multi-dimensional context information and the domain knowledge fragments are used as cue words to guide and constrain the large language model, and a workflow definition model is generated; and finally, a workflow engine is adopted to create the target workflow based on the workflow definition model, so that the processing accuracy of a natural language automatically generated by the workflow can be improved, the workflow is highly matched with actual business requirements of an enterprise, and the method can be applied to the technical field of business process intelligent management.
Owner:GUANGZHOU CHILDRENS VISA TECHNOLOGY CO LTD

Device for signing and extracting original text based on large model generation result comparison

The invention discloses a device for signing and extracting an original text based on a large model generation result, and the device is characterized in that a knowledge base management module manages an uploaded file, and analyzes and converts the file into an OFD file; the reasoning question-answering module selects a question-answering model to set a corresponding question-answering prompt word, returns reasoning content and records a file source and an address corresponding to the reasoning content; the original text searching module reads the reasoning result and the original text reference; the diversified endorsement module combines and displays the generated file and the original text, and carries out diversified endorsement and rendering; the combination file generation module synthesizes the signed files to form an OFD file combination file book with the signing effect, and meanwhile, the access of a third-party personnel organization structure is supported, and the files are audited. Based on an artificial intelligence large model and a private knowledge base, a user can initiate services fundamentally, and the user experience is improved. And the knowledge base and the business initiation content are quickly combined by adopting a file combination technology, so that the file organization time of the user is shortened.
Owner:JIANGSU ZHONGWEI TECH SOFTWARE SYST

Method and device for realizing private knowledge model and internet data management in large model environment, processor and storage medium thereof

The invention relates to a method for realizing private knowledge model and internet data management in a large model environment, which comprises the following steps of: acquiring a public knowledge model data set in multiple channels, storing knowledge data in the form of pdf, txt or json files, establishing an FAISS vector database, and cleaning data at the same time; deploying an open source large model, and performing fine adjustment on parameter factors and hyper-parameters on the model; combining a private knowledge vector database with the open source large model to construct a private knowledge model; performing a retrieval test on the private knowledge model in a private knowledge scene; and performing efficient indexing, data fusion, supplementary retrieval and data operation and management processing on the data input by the user. The invention further relates to a corresponding device, a processor and a storage medium thereof. By adopting the method, the device, the processor and the storage medium of the invention, multiple advantages of efficient data management and utilization, data security enhancement, knowledge management capability improvement, user experience optimization and the like are realized.
Owner:PRIMETON INFORMATION TECH

An agent cross-domain knowledge hybrid retrieval method and system for privacy protection

This invention provides a privacy-preserving intelligent agent cross-domain knowledge hybrid retrieval method and system, relating to the fields of artificial intelligence and data security technology. The method includes: responding to a query request initiated by a cloud-based intelligent agent, parsing the privacy tags embedded in the request workflow definition; and concurrently executing the following two operations through an asynchronous task scheduler: sending the public subquery to a cloud-based knowledge retrieval service, and routing the private subquery to a privacy computing agent client deployed in the target local environment; the privacy computing agent client executing the private subquery in a local private knowledge base; and at a cloud aggregation point, aggregating the decrypted ciphertext knowledge fragment set and the public knowledge fragment set into unified knowledge data, and returning it to the intelligent agent. This invention achieves efficient, secure, and integrated retrieval and utilization of distributed heterogeneous knowledge bases by a cloud-based intelligent agent while ensuring that the original privacy data does not leave the local security boundary.
Owner:SHANDONG GUOSHU DEV CO LTD +1

Domain Specific Agents for an AI Agentic Architecture

A method of selecting and interacting with agent types for an agentic AI system includes deploying an agent based on a template, with the agent assigned to a functional domain required by an enterprise, wherein the agent is selected to be one of a Prescriptive Conversational Agent; a Prescriptive Action Agent; a Q&A Retrieval Agent for private knowledge bases; a Q&A Retrieval Agent for public knowledge bases; a Dynamic Multi-task Action Agent; and a Personal Assistant Agent. The selected agents can be connected to a reasoning engine and orchestration module able to send and receive requests to the agent.
Owner:AUTOMATION ANYWHERE INC

Test case management method and device, equipment, medium and product

The invention discloses a test case management method and device, equipment, a medium and a product. The method comprises the following steps: analyzing historical test data through a large model and a fault mode recognition model to obtain a private knowledge base of the large model; the private knowledge base comprises a base knowledge base and a dynamic association rule; based on user test requirements and a private knowledge base, test parameters meeting the user test requirements are determined through a large model; and generating a target SQL test case meeting the test requirements of the user according to the test parameters. According to the embodiment of the invention, the coverage rate and accuracy of the test case can be improved.
Owner:JINZHUAN INFORMATION TECHNOLOGY CO LTD

A personalized practical training guidance and evaluation method and system based on a private knowledge base

The application relates to the field of artificial intelligence and industrial internet technology, in particular to a personalized practical training guidance and evaluation method and system based on a private knowledge base; the method comprises the following steps: acquiring user feature data, mapping a practical training task configuration file; generating guidance instructions and feeding back to an operation terminal; collecting and recording time sequence operation data; after the practical training task is completed, multi-dimensional comparison and analysis are carried out, and an evaluation report containing task completion quantitative data, expert operation model deviation degree analysis and parameter correction suggestions is generated; the application provides a personalized practical training guidance and evaluation method and system based on a private knowledge base, and solves the technical problems of lack of adaptability in practical training scheme configuration, lack of safe and controllable real-time guidance, and lack of objective quantitative means in practical skill evaluation in the prior art.
Owner:HANGZHOU ZHIJUAN TECH CO LTD

Cross-language federated knowledge graph embedding method for privacy protection

The invention discloses a cross-language federated knowledge graph embedding method for privacy protection, which belongs to the field of federated knowledge graph embedding and cryptography and comprises an initialization stage, a local training stage, an intra-group knowledge sharing stage and an inter-group knowledge alignment stage. The participants mainly comprise a knowledge graph client, a computing server and an assistant, the assistant is responsible for generating a secret key and a random number, the knowledge graph client uses private knowledge graph training locally to obtain a knowledge graph embedded model, an embedded vector of an aligned entity / relationship is selected according to a partial alignment set, and the embedded vector is encrypted and uploaded; and the computing server performs aggregation operation on the embedded vectors of the same entity / relationship and returns the aggregation operation to a corresponding client, the client verifies an aggregation result at a certain probability, and the operation is repeated until a specified number of rounds is reached. According to the method, cross-language federated knowledge graph embedding can be carried out on the premise of privacy protection, and cheating behaviors of the server are defended.
Owner:SOUTH CHINA UNIV OF TECH

A management method of a multi-modal enterprise knowledge base system

The application provides a management method of a multimodal enterprise knowledge base system, comprising: integrating multimodal data in an enterprise, constructing a private knowledge base, and combining the private knowledge base with a public knowledge base; performing intention analysis on multimodal retrieval requirements submitted by a user based on a combination result, performing conditional retrieval on the private knowledge base and the public knowledge base based on the intention analysis result, and outputting multimodal query data; performing quality inspection on the conditional retrieval process, and performing dynamic optimization on the combination result based on the quality inspection result to obtain a final multimodal enterprise knowledge base, and performing deployment management on the multimodal enterprise knowledge base. The management effect of multimodal knowledge in the enterprise is greatly improved, and the retrieval accuracy of the knowledge is also improved.
Owner:SHENZHEN YUNZHIYIN TECH CO LTD

Knowledge base construction method and question and answer dialogue method and system based on generative large language model

The application relates to the technical field of data processing, in particular to a knowledge base construction method and a question and answer dialogue method and system based on a generative large language model. A knowledge base name and a vector database table storage structure for storing knowledge base document data in the knowledge base are created according to a question and answer dialogue business target demand, and knowledge base metadata information is written into the vector database. Document file Chinese text information corresponding to the question and answer dialogue business target demand is acquired. The text is segmented according to a text segmentation length and a text terminator, and a plurality of text blocks corresponding to the text information are obtained. A pre-trained large language model is used to extract a text feature embedding vector corresponding to each text block, and text block metadata information and the text feature embedding vector corresponding to the text block are written into the library table storage structure in the vector database. The data stored in the vector database is used as a knowledge base required by the question and answer dialogue business target demand. By constructing a private knowledge base and understanding and summarizing user questions and knowledge base contents based on a generative large language model, text dialogue is realized, so that the user experience is improved.
Owner:数字郑州科技有限公司

Language model training method and device, storage medium and computer program product

The invention discloses a language model training method and device, a storage medium and a computer program product, and relates to the technical field of natural language process.The method comprises the steps that after a second device trains a small language model, text data are input into the small language model to obtain a second prediction result; the first device inputs the text data into the large language model to obtain a first prediction result, calculates first knowledge distillation loss and optimizes the large language model based on the aligned first prediction result and second prediction result, and inputs the text data into the large language model to obtain a third prediction result; the second equipment calculates second knowledge distillation loss and optimizes the small language model based on the aligned second prediction result and third prediction result; and obtaining a trained large language model until a preset training ending condition is met. According to the method, under the condition that enterprise computing power resources are limited, the big language model absorbs privacy knowledge of the field where each enterprise is located, and meanwhile, the data privacy of each enterprise is also protected.
Owner:WEBANK (CHINA)

Low-rank adaptive parameter selective updating method and system based on activation variance awareness

The application relates to the technical field of parameter efficient fine-tuning (PEFT) and large language model, and particularly relates to a low-rank adaptive parameter selective updating method and system based on activation variance perception. l The overall change range of the output activation of each sub-module of the layer is depicted under the distribution of enterprise private knowledge documents; the activation variance estimation values of each sub-module on the corresponding input sample are arranged in descending order, a threshold value is determined by calculating a quantile, and subset division is performed; the LoRA adapter is uniformly initialized for all sub-modules, and a selective back propagation constraint strategy is used to make the model concentrate the computing resources into the proportion parameter space which is most sensitive to the change of the enterprise private knowledge. The application can finely control the parameter updating process in a data-driven manner without changing the overall structure of the original pre-training model and introducing additional inference delay, and has good engineering realizability and universality across models and tasks.
Owner:BEIJING LANGUAGE AND CULTURE UNIVERSITY

Dynamic knowledge update and privacy-preserving inference method based on search enhancement generation

PendingCN122509300ARealize real-time dynamic updateensure safetyLinguistic modelTheoretical computer science
The application provides a dynamic knowledge updating and privacy protection reasoning method based on retrieval enhancement generation, and relates to the technical field of artificial intelligence, which comprises the following steps: receiving a user request and constructing context data, updating and storing public and private knowledge bases; encoding the context data to obtain a context vector, and performing similarity retrieval in the public and private knowledge bases respectively to obtain public and private knowledge sets; performing privacy protection on the private knowledge set, mapping the private knowledge to semantic agent Token, calculating the weight coefficient of the semantic agent Token based on the correlation degree of the context vector and the private knowledge, and generating a semantic agent Token set; introducing reasoning constraints and templates, constructing a structured prompt and inputting a large language model, guiding the model to complete reasoning and generate a reasoning result, collecting feedback information and adaptively optimizing the model, thereby realizing dynamic updating of the knowledge base, ensuring the security of private data, and improving the accuracy and timeliness of large model reasoning.
Owner:NANJING INTERCONNECT INTELLIGENT TECH CO LTD

Indoor accompanying method based on low-cost terminal grid coding dynamic knowledge base

The invention discloses an indoor accompanying method based on a low-cost terminal grid coding dynamic knowledge base, and belongs to the technical field of indoor intelligent navigation, and the method comprises the following steps: S1, generating a reference station set according to an electronic map of an indoor area; s2, generating a grid set covered by each reference station according to the coverage areas of the reference stations in the reference station set; s3, loading private knowledge bases to the reference stations according to the grid sets covered by the reference stations; and S4, training the single-layer lightweight neural network according to the private knowledge base loaded in the reference station, and reasoning by using the trained lightweight neural network to generate final navigation information.
Owner:CHENGDU ZHONGAO TECHNOLOGY CO LTD

Data processing method and apparatus, storage medium, and device

PendingCN123115428ALinguistic modelProcessing
The method comprises: acquiring user feedback text data to be analyzed; performing retrieval on the user feedback text data based on a pre-constructed domain private knowledge base to obtain a candidate set of domain knowledge associated with the user feedback text data; inputting the user feedback text data and the candidate set of domain knowledge into a large language model for multi-dimensional information extraction processing to obtain a plurality of candidate viewpoints; performing vector similarity-based aggregation processing on the plurality of candidate viewpoints to classify the plurality of candidate viewpoints into existing viewpoint clusters or determine whether clustering is triggered; for similar viewpoint groups triggering clustering, calling the large language model to remove viewpoints with inconsistent semantics in the similar viewpoint groups, and generating a core representative viewpoint based on the removed viewpoints; and creating a new viewpoint cluster record based on the core representative viewpoint to improve the accuracy and aggregation efficiency of viewpoint analysis.
Owner:NETEASE (HANGZHOU) NETWORK CO LTD

Answer generation method, apparatus, computer equipment, and storage medium based on cloud disk knowledge base

PendingCN123115328APersonalizationUser input
This disclosure relates to a method, apparatus, computer device, and storage medium for generating answers based on a cloud-based knowledge base. It includes: responding to user-inputted text information and historical question-and-answer information; determining the knowledge domain associated with the text information; retrieving a target file associated with the knowledge domain from the user's cloud-based knowledge base; determining a general knowledge expansion intent based on the similarity between words in the text information and historical question-and-answer information and preset intent expansion words; and retrieving a general knowledge file from the cloud-based knowledge base when the general knowledge expansion intent exceeds a preset expansion threshold; and inputting the target file, general knowledge file, text information, and historical question-and-answer information into a pre-constructed response text generation model to obtain a response text for the text information. Therefore, the embodiments of this disclosure can effectively avoid the problems of general large-scale model answers lacking personalization and failing to access users' private knowledge.
Owner:CHINA MOBILE INTERNET CO LTD +1

Question and answer method and equipment

The invention relates to the technical field of artificial intelligence, in particular to a question and answer method and device. A question to be replied input by an account is received, a fusion feature vector corresponding to information concerned by the account and a private knowledge graph describing preference, habits or concerned information of the account are obtained, and each relation edge in the private knowledge graph further corresponds to an activation weight; the activation weight is used for describing the timeliness of the information described by the corresponding triad, and screening each triad in the private knowledge graph based on each activation weight to obtain a target triad meeting the timeliness requirement, and inputting information described by the to-be-replied question, the fusion feature vector and the target triple into the large model subjected to fine tuning training to obtain reply information. When the large model answers the to-be-answered question, information concerned by the account is considered, information such as preference and habits of the account is also considered, and targeted reply and personalized service of the large model are realized.
Owner:WEBANK (CHINA)

Operation and maintenance assistant construction method based on local large language model and block chain technology

The invention discloses an operation and maintenance assistant construction method based on a local large language model and a block chain technology, and belongs to the block chain technology, and the method comprises the following steps: constructing an operation and maintenance service assistant front-end system based on the local large language model; uploading the technical document to the operation and maintenance service assistant front-end system, and generating different private operation and maintenance knowledge bases; constructing a knowledge chain, and uploading the private operation and maintenance knowledge base to different block chains; and establishing a relay chain to realize data sharing among different block chain private operation and maintenance knowledge bases. According to the method, the operation and maintenance assistant is constructed based on the local DeepSeek large language model, the private knowledge base is constructed by utilizing the capability of the existing large language model, the vertical field capability of the model is improved, meanwhile, the private knowledge base is linked, and a cross-chain communication method is provided, so that the security and confidentiality of related key technical knowledge are guaranteed.
Owner:KUNMING LOGAN KSEC AIRPORT LOGISTICS SYST COMPANY

A boundary control method and system for intelligent dialogue management

The present invention discloses a boundary control method and system for intelligent dialogue management, which includes: obtaining and parsing a description file related to making dialogue scene management; preprocessing the parsing result; using the processed data as training data to train a large model to obtain an industry large model with good evaluation performance; constructing a corresponding prompt scene template according to the usage scenario corresponding to the trained industry large model; matching the corresponding prompt scene template according to the user's question query during the dialogue interaction process, and receiving the user's modification feedback result of the prompt scene template. The method preprocesses the description file and constructs a private knowledge base; by adding a private knowledge base to the large model, multiple rounds of dialogue with the user are realized in complex scenarios; it has the ability to solve complex reasoning problems; even if errors occur, effective error correction can be achieved through memory and feedback mechanisms; so as to improve the user's interactive experience.
Owner:SHANGHAI YANSHU COMPUTER TECH CO LTD

Drilling simulator-oriented intelligent teaching assistant system base LLM training method

The application provides a drilling simulator-oriented intelligent teaching assistant system base LLM training method, and belongs to the technical field of large language model training of oil drilling, and the method comprises the following steps: based on the intelligent teaching assistant system base large language model, drilling simulator simulation data is acquired, and field self-adaption pre-training is carried out; low-rank self-adaption is utilized to fine-tune the original weight, and operation steps and fault diagnosis generation are carried out; by constructing an enterprise private knowledge base, a large language model is trained by using retrieval enhancement, and a trained verification head is obtained through explicit supervision; a direct preference optimization loss function is defined, a retrieval-aware direct preference optimization loss function is obtained by combining the verification head, a large language model is optimized, and the large language model training is completed; the intelligent assistant framework capable of fusing external professional knowledge base, understanding equipment operation mechanism and having teaching guidance and error diagnosis capability is obtained, and the intelligent level of the drilling simulator is improved.
Owner:SOUTHWEST PETROLEUM UNIV +1

Industry scenario-oriented large model empowerment private knowledge base construction and integration method

The application relates to the technical field of large model industry application, in particular to a large model empowerment private knowledge base construction and integration method for industry scenarios, which comprises the following steps: multi-source heterogeneous data structured preprocessing, industry large model entity relationship extraction and knowledge graph construction, unified semantic space vector mapping, bidirectional knowledge alignment, privacy isolation knowledge integration interface layer construction, and collaborative reasoning scheduling whole process design; a large model fine-tuning, full life cycle incremental updating and hierarchical permission control mechanism are matched, high-precision construction of an industry private knowledge base and deep safety integration of a large model are realized, large model reasoning illusion is greatly reduced, response efficiency and business matching degree of a scenario reasoning are improved, and the method has good industry expansibility and engineering landing practicability.
Owner:CHINA NAT BUILDING MATERIALS TECH CO LTD +2

General-purpose model-based and transfer learning-based private knowledge content generation method

The application relates to a private knowledge content generation method based on a general large model and transfer learning. The method comprises the following steps: acquiring a structured database, performing foreign key association between tables on structured relational data in the structured database to obtain table association data; processing the table association data to obtain a text data set and an instruction data set, respectively, performing lora fine-tuning on the general large model by using the text data set and the instruction data set to obtain a private knowledge large model; acquiring real-time private data, constructing a reference material vector library according to a plurality of segmented text vectors corresponding to the real-time private data, selecting a reference material vector closest to an input vector corresponding to user input from the reference material vector library to obtain corresponding reference material; and splicing the user input and the reference material according to an instruction template, inputting the spliced result into the private knowledge large model to obtain corresponding generated content. The method can improve the real-time performance and effectiveness of the generated content.
Owner:NAT UNIV OF DEFENSE TECH

Intelligent tourism service method based on structured management and knowledge enhanced questions and answers

The invention discloses an intelligent tourism service method based on structured management and knowledge enhanced questions and answers, and relates to the technical field of artificial intelligence and tourism information services, and the method comprises the steps: generating a structured semantic fragment through context fusion and anaphora resolution, carrying out the multi-dimensional vector matching of the structured semantic fragment and a private knowledge base, and calculating a knowledge correlation index Zst; and meanwhile, a scene vector weighted score Cjq is utilized to activate a scene Prompt template, and a scene fitness index Cjs is calculated, so that deep understanding of the continuous intention is realized. After a text is generated, decision keywords are extracted, a multi-modal correlation index Dmt is calculated through a cross-modal similarity model, and materials are screened according to a threshold value. And finally, performing weighted fitting on the knowledge relevancy index Zst, the scene fitness index Cjs and the multi-modal relevancy index Dmt through a preset weight coefficient to obtain a comprehensive quality index Zhpz, so as to screen and output an optimal scheme of combining a text and a multi-modal material, thereby realizing an accurate, coherent and attractive intelligent question and answer service.
Owner:XIAMEN XIAOTI TRAVEL TECH CO LTD

Information generation method and device, equipment, storage medium and computer program product

The invention relates to the field of natural language processing, and discloses an information generation method, device and equipment, a storage medium and a computer program product.The method comprises the steps that in response to input content of a large model, retrieval is carried out in a preset private knowledge base according to the input content, a retrieval result is obtained, and based on the input content, the retrieval result is used as a generation reference, and the large model is generated; response information is generated through the large model; according to the method and the device, the knowledge in the preset private knowledge base is retrieved to enable the large model to refer to answer, so that the large model has the capability of answering private questions, the defect that the large model is trained by public corpora and cannot answer private knowledge related questions in the prior art is overcome, and the answering accuracy of the large model is improved.
Owner:BEIJING QIYUAN TECH CO LTD

Method and system for multi-user knowledge interconnection and intelligent representative interaction

The invention relates to the technical field of knowledge social interaction, and discloses a method and system for multi-user knowledge interconnection and intelligent representative interaction, and the method comprises the steps: relying on a large language model and a retrieval enhancement generation technology, analyzing the multi-modal content of a user, extracting a structured knowledge unit, and generating a multi-dimensional knowledge representation, constructing a personal knowledge portrait and storing the personal knowledge portrait into a private knowledge base; a personal intelligent representative is constructed, and cross-user knowledge accurate retrieval and authority control are realized; dynamically calculating the credibility level of user knowledge, and driving the self-adaptive adjustment of the authority; social behaviors are converted into knowledge graph optimization signals based on reinforcement learning, a knowledge interaction closed loop is constructed, and high-value knowledge is precipitated; the right is confirmed and the knowledge value is quantified through a double-track recording mechanism, the income is distributed based on a Shapley value algorithm, and a credible settlement basis is generated. The problems that existing platform knowledge and identity binding is loose, credibility is difficult to guarantee, and interaction is low in efficiency can be solved.
Owner:BRICS FUTURE NETWORK RES INST (SHENZHEN CHINA)