77results about How to "Improve professionalism" patented technology

The invention discloses an industrial measurement data intelligent analysis report generation method and system based on a large language model, and relates to the technical field of industrial measurement, and the method comprises a multi-source data and abstraction module which carries out the structural processing of original data DAT output by measurement analysis software; and the large language model service and interface module is used for submitting the PCT to a selected large language model LLM through a multi-model adaptation interface to execute semantic reasoning and output a structured text TXT. Structured analysis and semantic association modeling are carried out on original measurement data through multi-source data and an abstract module, when overall alignment deviation or local feature anomaly exists in the original data, a hierarchical dependency relationship between the data can be established through a semantic graph structure, a deviation transmission path is revealed, and the accuracy of the measurement data is improved. The large language model is analyzed in a unified data context, and the utilization depth of measurement data can be effectively improved, so that the accuracy and integrity of report analysis are improved.

The invention discloses an LLM (Logical Language Modelica) domain method for Modelica intelligent modeling optimization, and belongs to the field of computer-aided modeling. Structured information is converted and extracted by collecting field data of the new energy automobile; then, based on the RAG technology, intelligent blocking and vectorization processing is carried out on the structured information to form a structured knowledge base, and then a recall test is carried out by calculating a comprehensive weighted score to judge whether the structured knowledge base is available or not; then, the LLM is integrated into an AI-Agent framework, an AI-Agent is formed, and LLM parameters are configured; integrating the structured knowledge base which passes the test into the AI-Agent by utilizing an API (Application Program Interface) and an access protocol; and meanwhile, a specialized cue word template is designed and integrated into the AI-Agent, and a Modelica code generation template and related constraints are standardized. And finally, the user asks a question to the AI-Agent, the LLM works according to a specialized cue word instruction, an API interface is called to access the knowledge base, and Modelica codes which conform to language specifications and can be operated by the user are automatically output through the LLM. According to the invention, end-to-end conversion from unstructured input to high-fidelity model output is realized.

The invention relates to a multi-agent cooperation-based language model reasoning resource scheduling method and system, and the method achieves the intelligent simulation of multi-view iterative thinking in a complex reasoning task through the construction of a multi-agent system which is clear in division of labor and is provided with a special knowledge base. View limitation and decision deviation of a single model in long-sequence and multi-step reasoning are effectively overcome; depending on the fusion of the general capability of the large language model and the task special knowledge base, the professionality and accuracy of the reasoning result are improved. Meanwhile, the problems of resource waste, redundant calculation and unstable convergence caused by cognitive overload in the cooperation process are solved by combining multi-round dynamic cooperation with a convergence mechanism of cognitive load perception and a dynamic token number limitation and low-confidence branch pruning strategy; therefore, on the premise that the reasoning quality is guaranteed, the utilization efficiency of computing resources is remarkably optimized, peak value occupation is reduced, the overall reasoning time delay is shortened, and reliable technical support is provided for efficient and stable deployment of a large language model in a complex task.

The application provides a multi-agent collaborative management method and system for whole-process engineering consultation, which comprises the following steps: S1, obtaining a whole-process engineering consultation task; S2, performing task decomposition on the obtained whole-process engineering consultation task based on an LLM large language model to obtain task decomposition features of the whole-process engineering consultation task, wherein the task decomposition features comprise one or more subtasks; S3, respectively performing task classification feature extraction on each subtask of the task decomposition features to obtain a task classification label corresponding to each subtask; S4, according to the task classification label corresponding to each subtask, matching a corresponding agent from candidate agents to process the subtask, and obtaining a subtask solution output by the agent; and S5, performing optimization processing according to the obtained subtask solution to obtain a whole-process engineering consultation task solution. The application helps to improve the effect of the whole-process engineering consultation task solution.

The present application relates to the technical field of stage lighting, and discloses an effect device for shaping stage light effect, comprising a lamp body, the lamp body comprising a connecting top seat and a lamp shell body, a light source is arranged in the lamp shell body, an adjusting device is connected to the lamp shell body, the adjusting device comprising: an extension shell body; two fixed pipes symmetrically connected to the extension shell body; an optical lens, both ends of the optical lens being connected to an upper cavity and a lower cavity respectively, the optical lens being a hollow structure, and the inner walls of the upper cavity and the lower cavity being fixedly connected to impellers; a connecting pipe, respectively connecting the upper cavity and the lower cavity; wherein, one side end face of the optical lens is a one-way perspective glass, the other side end face is ordinary glass, two fixed pipes are respectively connected to a liquid pipe, the two liquid pipes are both connected to an external water tank, a water pump is connected to the two liquid pipes, and the two liquid pipes are arranged on the inner wall of the lamp shell body, solving the problems of high energy consumption and single effect of the existing equipment.

This invention belongs to the field of automatic control technology and discloses a real-time linkage control system for multimodal devices in a conference room. It includes a multimodal device group, a control server, and an environmental sensor group. The system solves the problem of inter-device coordination conflicts in traditional conference systems by constructing a device execution conflict model. This model, based on a resource competition conflict matrix and action mutual exclusion conflict vectors, achieves accurate identification and prediction of device operation conflicts. Control parameters are optimized to give the system environmental adaptability. A closed-loop control loop enables coordinated device operation and dynamic parameter optimization, and closed-loop feedback adjustment can evaluate and correct linkage deviations in real time. During execution, the system achieves seamless collaboration among multiple devices through techniques such as calculating operating margins and adjusting time offsets, effectively improving the intelligence level of conference room environmental control.

This invention provides a method and system for semantic novelty search of scientific and technological literature based on a large-scale pre-trained model, belonging to the field of information retrieval technology. The method includes: acquiring the target user's query input text and retrieval constraints; performing preprocessing and standardization to obtain preprocessed query input text; performing multi-granularity semantic encoding fusion to determine the semantic vector of the preprocessed query input text; performing domain-adaptive fine-tuning to obtain an optimized BERT model; constructing a feature classifier to generate feature semantic representations to obtain feature semantic representation results; performing comprehensive novelty assessment to obtain a comprehensive novelty rating; and performing automated difference analysis to obtain difference analysis results. This invention solves the technical problem in existing retrieval schemes based on pre-trained models that use single-granularity semantic representations, which cannot fully capture the complex semantic structure of professional terms, resulting in inaccurate novelty search results.

The invention relates to an intelligent low-altitude economic scene passenger emotion pacifying method and system based on a large model. The method comprises the following steps: constructing a multi-modal emotional state data set of low-altitude travel passengers; wherein the multi-mode emotional state data set comprises passenger face video sequence data, in-cabin voice and audio data, heart rate variability time sequence data, respiratory frequency time sequence data and external situation context data; performing emotional depth feature extraction based on multi-source space-time signals on the multi-modal emotional state data set to obtain multi-source emotional features; emotion recognition is carried out based on the multi-source emotion features; based on the emotion recognition result, constructing a pacifying strategy generation model; and utilizing a pacifying strategy generation model to intelligently pacify the emotion of the passenger, and evaluating and feeding back the pacifying effect in real time. According to the emotion and specific inducements recognized in real time, pacifying measure suggestions which conform to safety specifications, fit scenes and are natural in language can be generated, and timeliness and specialty of passenger cabin service are improved.

The large model interaction method comprises the following steps: obtaining a user ocean query; inputting the ocean geographic query of the user into the ocean semantic analysis model to obtain query corpus features; matching the query corpus features with the grid hydrological database, and generating answers based on the ocean query of the user; by adopting the technical scheme, the accuracy and specialty of semantic analysis are improved, core information such as geographic expression, industry exclusive requirements and time-space constraints in user query can be effectively captured, the problem that a traditional model cannot answer questions due to semantic understanding deviation is fundamentally avoided, and the user experience is improved. It is ensured that the matched data is highly matched with user requirements in the geographic range, the time window and the industry attributes, and then answers conforming to the geographic characteristics of the ocean scene are accurately output.

The invention relates to a content generation method combining a knowledge graph and a large language model. The method comprises the following steps: based on multi-type target content data, extracting an entity, relationship and attribute triple of a multi-source corpus, dividing a theme cluster, constructing a knowledge graph, and carrying out standardized packaging on knowledge units to obtain a standardized knowledge set; analyzing a user input demand, obtaining a multi-dimensional demand feature, performing similarity matching with a theme cluster to obtain a target cluster, screening candidate knowledge through multi-hop retrieval, and obtaining an associated knowledge subset through three-dimensional evaluation index empowerment score sorting; converting the associated knowledge subset into an analyzable structured text, and adjusting and optimizing model generation parameters to obtain initial generation content; and comparing the initial content with a knowledge graph to obtain a result, optimizing text features and performing adaptive adjustment in combination with multi-type target content data and audience features, and obtaining standard content of target quality. Content generation in combination with the knowledge graph and the large language model is realized.

This invention relates to the field of talent training quality assessment technology in medical and pharmaceutical colleges, and discloses a big data closed-loop assessment system for talent training quality in medical and pharmaceutical colleges. The system includes: a data acquisition module to collect multi-source training data; a preprocessing and fusion module to construct a talent training quality dataset; a big data closed-loop assessment module containing three-level indicator construction, multi-dimensional assessment, and result analysis units to complete the assessment and generate personalized reports; an intelligent feedback module to push and visualize the assessment results; a training optimization module to adapt rigid constraints and transform them into optimization measures; and a system iteration module to ensure the assessment system is updated synchronously with the pharmaceutical industry. This invention constructs a full-link closed loop, solving problems such as assessment disconnect and low accuracy, improving the quality of talent training and its industry adaptability, and adapting to the characteristics of medical and pharmaceutical college training.

The utility model discloses a filter support, including fixed frame, two active elements, two driven module and transmission part. Active element and driven module all are opposite fixed frame mobile setting, and transmission part is opposite fixed frame rotation setting. Two active elements and two driven module are along annular arrangement, and a driven module is located between two active elements. Active element and transmission part transmission connection, and driven module also with transmission part transmission connection. When active element is opposite fixed frame mobile, can drive transmission part opposite fixed frame rotation, and then promotes driven module to move to the same mobile direction with active element. The structure realizes the cooperation of active element and driven module through transmission part, improves the adjustment flexibility, can better adapt to different caliber lens, reduces the dark corner, satisfies the diversified shooting demand.

The invention discloses a curriculum commodity personalized recommendation method and device, a terminal and a storage medium, and relates to the field of commodity recommendation. The method comprises the following steps: acquiring client target data and multi-dimensional current situation data to perform state quantification to obtain a quantized state set, constructing a course logic relation graph, mapping knowledge level differences to graph effective nodes, performing topological sorting and time resource filtering to obtain candidate learning paths, and performing learning on the candidate learning paths. And calculating the comprehensive gain value of the candidate learning path, and selecting the path corresponding to the highest value as a personalized recommendation path. Meanwhile, the method also comprises a rapid recommendation step for new customers or data sparse customers. The technical effects of providing a personalized course commodity recommendation scheme according to the actual condition of the customer, improving the accuracy and effectiveness of recommendation, saving the learning time for the customer, improving the learning effect and facilitating the new customer to quickly obtain proper recommendation are achieved.

The invention discloses a water conservancy knowledge base and intelligent question and answer construction method based on a large model and RAG, and the method comprises the steps: 1, combing a water conservancy knowledge system, and sorting water conservancy professional documents and data as materials of the knowledge base; step 2, constructing a water conservancy knowledge library management system through a J2EE system, a B / S architecture and a distributed object storage system; step 3, based on the RAG technology, forming different knowledge bases according to the sorted knowledge classification and confidentiality security level of the water conservancy industry, and establishing a water conservancy knowledge intelligent question-answering system on the knowledge bases; 4, performing extended application on the basis of the constructed water conservancy knowledge intelligent question-answering system; the problems that in the prior art, due to the fact that unstructured data processing is weak when a water conservancy knowledge base is constructed, a large number of image-text mixed documents such as PDF and PPT cannot be effectively utilized, and the recall rate and the accuracy rate are not high are solved.

The present application belongs to the technical field of nuclear power plant physical simulation and simulation, and particularly relates to a nuclear power plant core instrument bay equipment simulation body and a simulation method. The simulation body comprises a driving unit, a group selector assembly and a path selector assembly, the driving unit is connected to the path selector assembly through the group selector assembly, the cable outlet end of the driving unit is connected to the group selector inlet end through a stainless steel pipe and a sleeve joint, and the group selector outlet end of the group selector assembly is connected to the path selector inlet end of the path selector assembly through a stainless steel pipe and a sleeve joint. The present application has the beneficial effect of improving training effect and efficiency, the RIC system simulation body can simulate the real operating environment of the nuclear power plant core, so that the trainees can learn and operate in a work scene close to the actual one, thereby more intuitively understanding the working principle and operation process of the core measurement system.

The application discloses a retrieval enhancement generation method and related products, which comprises the following steps: obtaining a query vector; matching the query vector with a multi-dimensional feature graph to determine a target node; the multi-dimensional feature graph comprises multiple nodes and multiple edges; the node represents a node vector; the edge represents a relationship vector; there is an index relationship between the node vector and the relationship vector; the multiple nodes comprise the target node; performing feature extraction on the query vector to obtain a relationship feature corresponding to the query vector; performing dynamic graph walking in the multi-dimensional feature graph based on the target node and the relationship feature to obtain one or more effective paths; and obtaining a final answer based on the one or more effective paths. The application effectively avoids the "illusion" risk of a large model, ensures that the output content is compliant, reliable and auditable, improves the accuracy of the retrieval result in retrieval enhancement generation, and effectively breaks through the inherent limitations in semantic integrity, multi-hop reasoning capability and relationship intention understanding.

The invention discloses a supporting frame for computer hardware teaching demonstration, and relates to the field of teaching supporting frames, the supporting frame comprises a supporting seat and a supporting cylinder arranged in the supporting seat, the supporting cylinder is slidably provided with a bearing box, the bearing box is internally provided with an upper layer cavity and a lower layer cavity, and the upper layer cavity is internally provided with a placing box; a placing plate is arranged in the lower-layer cavity, a control assembly is arranged in the supporting cylinder, a handle is fixedly arranged on the supporting cylinder, a first unfolding assembly is arranged in the handle and the upper-layer cavity, and a second unfolding assembly is arranged in the lower-layer cavity. During demonstration, the bearing box ascends, the placement box and the placement plate are unfolded to form a three-dimensional display space, the space utilization rate is maximized, the placement plate is in an inclined state, students in the back row can easily and clearly see the whole view and layout of the plate surface, the problem that the sight is blocked during static flat placement is solved, and the teaching efficiency of teachers is improved.

The invention belongs to the technical field of chronic disease management, and discloses an intelligent chronic disease management method, system and device based on pre-analysis and map retrieval enhancement and a medium, and the method comprises the steps: firstly, achieving the dynamic evaluation and early warning of the health state of a user in combination with the physiological data of the user; secondly, designing a lightweight user demand pre-analysis mechanism, performing semantic analysis and structured reconstruction on user input, and combining health tags and physiological data to improve the understanding and adaptation ability of a large language model to individualized demands; and finally, by means of a chronic disease knowledge base and a map retrieval enhancement technology, deploying a domain model oriented to a specific disease scene so as to improve the vertical adaptation capability of the model. According to the technical scheme, a chronic disease knowledge service framework enhanced through pre-analysis and map retrieval is fused, the accuracy and normalization of knowledge services can be effectively improved, and an extensible technical path is provided for intelligent chronic disease management.

The invention discloses a large model training method and system applied to an education scene, and belongs to the technical field of large model training, and the method comprises the steps: inputting corpus data in the field of education knowledge into a pre-constructed base large model for incremental pre-training, and obtaining a target base large model; performing full fine tuning training on the target base large model to obtain a fine-tuned target base large model, performing reinforcement learning training on the fine-tuned target base large model, and then performing full fine tuning training on the learned target base large model; and performing data cleaning on the third fine adjustment corpus for four times, inputting the obtained fourth fine adjustment corpus into the target base large model after secondary fine adjustment, and performing reinforcement learning training to obtain a final target base large model. According to the method, heuristic answers are obtained by adopting a specific training mode, and training is performed by using different cleaning steps and different corpus data, so that the accuracy and specialty of a large model for data of different subjects in the education field after training are improved.

The invention discloses a method for risk management and control after a water transportation platform draws a freight in advance, and relates to the technical field of water transportation logistics finance, risk control is extended to be after pre-drawing, a comprehensive risk level is generated through continuous monitoring, fusion of repayment capability, operation stability and abnormal behavior three-dimensional scoring, dynamic quota adjustment is linked, and the risk management and control efficiency is improved. The method has the advantages that the risk control link is extended from only one-time auditing before loaning to after a ship owner takes the freight in advance, static evaluation is replaced by continuous monitoring, and the risk control efficiency is improved by combining the business characteristics of the water transportation industry. A multi-dimensional risk assessment system covering repayment behaviors, operation states and abnormal behaviors is constructed, hidden risk signals such as operation deterioration and repayment abnormity are actively captured instead of being passively handled after overdue or bad debts occur, and the defects that in the prior art, risk identification lags behind, and dynamic monitoring means after taking is lacked are effectively overcome.

The present application relates to the technical field of large model, especially to a reading recommendation large model generation method and system based on big data resources.The method comprises the following steps: obtaining a text to be evaluated and collecting reading materials of multiple learning stages and multiple types, labeling them, aggregating and verifying the consistency of the labeling results, and constructing a verified recommendation database; based on the recommendation database, performing low-rank adaptive fine-tuning on the general pre-training large model to obtain a domain model, and monitoring and correcting the performance and bias during the training process; performing multi-dimensional analysis on the text to be evaluated to generate a text feature vector, combining semantic vector similarity retrieval to perform hybrid retrieval on the recommendation database; the present application realizes more accurate and efficient reading recommendation through the construction of the reading recommendation large model.

The invention discloses an online inquiry processing method and device, electronic equipment and a computer readable medium, and relates to the technical field of online medical treatment, and the method comprises the steps: responding to a triggering operation of a user for an online consulting room control, and obtaining historical symptom consultation data of the user in an online consulting room within a preset time; in response to the situation that the historical symptom consultation data is not empty, generating a guide copywriting according to the historical symptom consultation data, and pushing the guide copywriting to the user side; receiving reply data for the guide copywriting returned by the user side, and performing triage according to the historical symptom consultation data and the reply data to determine a specialized large model; and under the condition that the preliminary diagnosis process is executed, the specialized large model carries out inquiry interaction with the user side based on a preset inquiry question, obtains answer data returned by the user side, generates a diagnosis and treatment suggestion based on the obtained archive data, chief complaint data and the answer data, and pushes the diagnosis and treatment suggestion to the user side. And the flexibility of an online inquiry process and the accuracy of online inquiry processing are improved.

The invention relates to the technical field of natural language processing, in particular to a teacher work document auxiliary generation method based on a large language model, which mainly comprises the following steps: constructing an expert model based on the large language model, and training the expert model through supervision and fine tuning to obtain a first pre-training model; obtaining pre-constructed tool calling data, and performing instruction fine tuning on the first pre-training model by using the tool calling data to obtain a pre-trained expert model; packaging a data acquisition interface of an external system, and pairing the data acquisition interface with a task in the tool calling data; and obtaining a demand instruction of a user, and inputting the demand instruction into the pre-trained expert model. The method can efficiently and automatically assist teachers in generating work documents by utilizing a fine-tuned large language model and a multi-module architecture according to data and scene characteristics in the education field.

This invention provides a multi-agent traceable decision-making system and method for environmental data mining, belonging to the interdisciplinary field of artificial intelligence and environmental informatics. The system includes: a task complexity assessment module, a multi-agent conditional collaboration module, an arbitrator triggering and adjudication module, an environmental knowledge graph module, a feedback learning optimization module, and a decision tracing and visualization module. The method includes the following steps: Step 1: Task complexity assessment and collaboration mode allocation; Step 2: Parallel processing and dynamic collaboration of two intelligent analysts; Step 3: Consistency check and arbitration triggering; Step 4: Arbitrator adjudication and weight integration; Step 5: Human expert correction and weight optimization feedback; Step 6: Decision tracing and weight visualization. This invention enables more accurate decision optimization, significantly improves the professionalism and accuracy of decision-making, and possesses the ability to proactively identify and prevent potential problems.

The invention relates to the technical field of computer data processing and natural language processing, in particular to a case clue identification and analysis system, which comprises a data access and positioning module for calling a preset multi-source map service interface to concurrently analyze a network address, and determining a target position through voting arbitration; the cascade semantic analysis module is used for identifying attributes by utilizing coarse granularity of a first-level classification model, and routing to a vertical expert agent to generate a fine-granularity label; the element feature extraction module is used for eliminating interference digital features and then generating standard amount features, and analyzing intention intensity of the user; the value quantification and pricing module is used for mapping the heterogeneous features into multi-dimensional scores based on a tree-shaped weight system, and generating clue commercial value scores and pricing results; according to the invention, the problem of value evaluation caused by information non-standardization in a compliance consultation scene is solved.

The invention provides a vector knowledge base construction method and system based on a large model and a retrieval method.The construction method comprises the following steps that an original document is segmented to obtain a plurality of text blocks, and metadata information of the text blocks is extracted; extracting a concept entity from the text block based on a large model to form an initial concept entity; performing semantic matching and integration on the initial concept entity based on a large model to obtain a standard concept entity; establishing an occurrence association relationship between the text block and the standard concept entity; based on the standard concept entity and the original text of the standard concept entity, utilizing a large model to analyze and construct an entity map representing the semantic relationship between the entities; after the text blocks and the standard concept entities are subjected to vectorization processing, metadata information of the text blocks, the incidence relation between the text blocks and the standard concept entities and the entity maps are correlated to form a vector knowledge base. According to the method, more accurate and more intelligent knowledge management and retrieval can be realized.

The application provides a three-dimensional garment design diagram generation method and device and electronic equipment, and belongs to the technical field of garment design. In the method, the first text information is taken as the input of a three-dimensional garment design diagram generation model, the first text information uses natural language to describe color information, style information, material information and detail information of a first three-dimensional garment design diagram expected by a user, the first text information is predicted and processed by the three-dimensional garment design diagram generation model, and a first three-dimensional garment design diagram is obtained, so that efficient conversion from natural language to a three-dimensional garment design diagram is realized. The three-dimensional garment design diagram generation model takes text information expressed in natural language as input, and significantly improves the convenience and flexibility of user expression of design intentions. Meanwhile, the first text information is used to accurately express garment color, style, material and details, effectively narrows the information gap between design intentions and generation results, and improves the authenticity of the three-dimensional garment design diagram.

The embodiment of the application discloses a kind of event risk detection methods, equipment and medium based on large model, belong to risk detection technical field, solve the problems that prior art scene poor adaptability, rely on sample and artificial rule, so that the risk detection accuracy and efficiency of complex and diverse event are lower. Including, based on the industry knowledge and industry scene information obtained, event risk detection processing template is constructed;Wherein, event risk detection processing template is used to provide detection processing logic for event risk detection model;Obtain the event basic data corresponding to the event to be detected, retrieve event related knowledge data in the preset vector database based on event basic data;Event basic data and event related knowledge data are filled into event risk detection processing template to obtain input prompt word;Call event risk detection model, based on event risk detection model and input prompt word carry out event risk detection, and output detection conclusion.

This invention relates to the field of systems engineering modeling and simulation technology, and discloses a semi-automatic MBSE intelligent modeling and simulation framework based on a large language model. This framework consists of four modules: semantic data acquisition, prompt word engineering, modeling and simulation code generation, and automated modeling and closed-loop simulation verification. The system parses modeling requirements and key trajectory elements into standardized control instructions, integrates structured scene data to form structured prompts, retrieves engineering code fragments and calls the large language model to generate modeling code and simulation scripts, runs simulations, and performs consistency verification according to preset parameter constraints, achieving end-to-end automation and traceability. The framework completes scene initialization and object instantiation through interface-driven processes and outputs scene object parameters through a unified data interface; it supports result feedback triggering re-retrieval and regeneration, reducing manual scripting and interface work, and improving the consistency, efficiency, and applicability of modeling and simulation.

The application discloses a kind of instruction fine-tuning data end-to-end synthesis system based on self-guided workflow, the system includes: task automatic discovery module, for based on preset prompt word and generation model, the original corpus data of target field is carried out semantic analysis, obtain multiple task modes in original corpus data, and generate instruction template for each task mode;Closed loop self-optimizing workflow module, through instruction data generation model, based on instruction template output multiple instruction data of target field;Use these data to carry out field fine-tuning and effect evaluation to target large language model;Then, according to the evaluation result, the preset prompt word is optimized, and the updated instruction template is obtained to drive the next round of iteration. Fine-tuned model can be used for the question and answer, abstract, reasoning, comparison and other tasks of target field.