10 results about "Design language" patented technology

The invention discloses a multi-language general part-of-speech recognition method and system based on a large language model, and relates to the technical field of natural language processing and part-of-speech recognition, and the method comprises the steps: employing the large language model as a core base model, and carrying out the efficient fine tuning, language design and multi-field vocabulary and syntactic analysis through a low-rank adaptation technology, model distillation is used for migrating knowledge of a large model to a small model so as to reduce the size and calculation cost of the model, standardized cue words are designed to guide the model to generate standardized output, a part-of-speech recognition task is automatically completed from the standardized output, and finally JSON format data are generated through code analysis model output. And therefore, the front-end display and the back-end service logic processing can be efficiently supported. The output part-of-speech information can be used for various downstream tasks such as intelligent search, text classification, code analysis and grammar verification, and data support is provided for various service systems.

The invention discloses a method and system for determining a cable tensioning construction scheme of a spoke type string structure, and the method comprises the steps: building an overall analysis model through ANAYS for the tensioning construction scheme of the spoke type string structure, and carrying out the parameterization of different tensioning schemes through compiling an ANSYS parameterization design language on this basis, batch circulation calculation is carried out in the tensioning process, and the calculation efficiency is high; and meanwhile, the advantages and disadvantages of different construction schemes can be quickly evaluated by comparing the response change rate of the structure in the tensioning construction process, so that the batch tensioning scheme of the spoke type string structure can be quickly determined.

The application discloses a method for cluttered scene target object grasping based on visual-language-action joint modeling. The application uses object-centered representation to realize a method for cluttered scene target object grasping based on visual-language-action joint modeling, processes the object-centered representation through a pre-trained visual-language model and a grasping model, obtains visual-language features and grasping features of each bounding box, and uses a transformer to implement cross-attention mechanisms among visual-language-action multimodality, generates visual-language-action cross-attention features, and then generates decisions and executes, so that higher sample utilization is realized, and additional data collection and training in the simulation-physical migration process are avoided; compared with a two-stage strategy, visual attributes and planner screening rules for language-visual matching need not be artificially designed, so that more flexible language instructions can be adapted, and better task generalization is achieved.

The present invention provides a method for modeling a multi-core processor based on AADL. The multi-core processor is modeled using the Architecture Analysis and Design Language AADL. Different types of AADL components such as system components, process components, memory components, device components, and abstract components are established to model the overall multi-core processor and various parts inside the multi-core processor, including the processor part, memory part, device part, cluster structure, etc. Interaction ports port, data components data, connections connection, etc. are established to ensure the type correctness and flow correctness of data transmission. Referring to the physical composition structure of the multi-core processor, the inclusion relationship and interaction connection relationship between components are checked, so as to use the established model to experiment with the capabilities of the multi-core processor on an open-source and free platform, achieving the purpose of liberating human resources and saving financial expenditures.

The invention provides a radar embedded software integration method based on model driving by aiming at the requirement of rapid and high-quality integration of large-scale complex radar embedded software and overcoming the existing software integration mode based on artificial coding, and through the technologies of model analysis, model mapping, code assembly and the like, the radar embedded software integration method based on model driving is realized. The software design model is automatically converted into the software code, efficient and high-quality integration of the software code is achieved, the software coding time is greatly shortened, and manpower and time cost is effectively saved. High-precision conversion from a design language XML to a code language is automatically completed through model mapping; according to code assembly, high-quality codes are automatically generated strictly according to software coding specifications and coding formats, the space resource allocation problem in radar embedded code engineering implementation is fully considered, and the real-time requirement of radar embedded software operation is met.

The invention discloses a low-code technology for generating micro-service based on a design mode. The technology comprises four parts of business design language, language translation, micro-service compiling and containerized micro-service operation. The method is characterized in that a business language of a business hierarchical structure is constructed based on a domain-driven design (DDD) idea, and description is carried out through an XML script convenient for visual expression; translating into a current popular target language for software development based on grammar analysis of a business language, forming a code project of the target language, and meanwhile, carrying out secondary modification on code project support; a micro-service business module capable of being operated is formed based on compilation of code engineering, and the low-code or code-free process from design to the module capable of being operated is achieved. According to the invention, a micro-service operation environment irrelevant to an operation system platform is established, and operation isolation and unified management of operation of micro-services are realized. The low-code technology follows the design rule and process of a traditional software project, is the optimization of the traditional code research and development process, is particularly suitable for research and development of an information management system, and is convenient for non-invasive integration of a historical legacy system and reduces the risk caused by reconstruction of historical legacy problems in addition to the micro-service characteristics of the information management system.

The application discloses an architecture design method and device of an intelligent automobile cyber-physical system and related equipment. It relates to the technical field of vehicles and road traffic. The method comprises the following steps: obtaining the architecture design requirements of the intelligent automobile cyber-physical system; in response to the architecture design requirements, constructing the view architecture of each perspective according to the metamodel of each view in the preset system architecture framework under multiple perspectives, and obtaining the initial architecture model of the intelligent automobile cyber-physical system; and simulating and optimizing the initial architecture model to obtain the target architecture model of the intelligent automobile cyber-physical system. According to the embodiment of the application, the design path and output standard are specified by a unified architecture framework, the design language and understanding, design boundary and analysis granularity of multiple participants are unified, and the design consensus of multiple stakeholders is reached.

This invention relates to a method and system for generating site-specific sustainable architectural design. The method first acquires architectural design text and case images to construct expected and unexpected sample sets. Then, a large language model is used to process the expected sample set to construct a knowledge graph for site-specific sustainable architectural design. Simultaneously, expert and public perceptions are extracted from the sample set to generate feature heatmaps. These heatmaps are then analyzed by the large language model with the knowledge graph to extract design elements and design language conducive to site-specific sustainable architectural design, constructing a knowledge base for site-specific sustainable architectural design. Subsequently, a LoRA model is trained based on this knowledge base, a supporting prompt word system is constructed, and an evaluation model is attached to create a knowledge-guided AIGC workflow for site-specific sustainable architectural design. Finally, the current status image of the target building is input into the workflow, and the output results are filtered to obtain the site-specific sustainable design scheme for the target building. Compared with existing technologies, this invention has advantages such as improved design generation efficiency and quality.

The invention discloses an aspect-level cross-language sentiment analysis framework, a product, a medium and equipment, and belongs to the technical field of artificial intelligence. The framework is an improved MIXER framework, and the improvement of a code hybrid context demonstration module comprises the following steps: constructing a multi-language data set based on a source language data set and a target language data set so as to enrich semantic information; semantic alignment optimization is carried out by introducing a similarity scoring mechanism, so that the model capture capability is improved; the improvement of the language specific tag module comprises the following steps: designing a language category tag for the constructed multi-language data set; taking the language category label as a part of model input, and guiding the model to pay attention to language differences; and the high-efficiency fine tuning module is used for finely tuning the optimized parameters so as to support high-efficiency adaptation of the whole frame on a large-scale pre-training model. According to the method, the performance of cross-language sentiment analysis can be improved, and the multi-language adaptive capacity of the model is enhanced while dependence on target language annotation data is reduced.

The invention discloses a process engine-based ship control system information layer software design method, which comprises the following steps of: 1) carrying out demand analysis, and determining a task process involved in a scene; 2) summarizing generality among task processes, and constructing a corresponding hierarchical model; and 3) flow node decomposition and combined construction: converting the abstract model into an executable flow. According to the method, control tasks, which need to be realized by ship control system information layer software, in different fields of the ship are described by adopting a universal task flow design language, so that control connection and linkage of the ship in different fields can be realized, and the communication cost of control linkage among different fields is reduced; and the development cost is also reduced by a graphical interface.