Method and device for generating attached drawings of patent application, storage medium and electronic equipment

Abstract

The invention discloses a method and device for generating an attached drawing of a patent application, a storage medium and electronic equipment, the method is based on a large language model, and the method comprises the following steps: obtaining text information of the patent application, the text information comprising claims and / or specifications of the patent application; analyzing internal logic in the text information, and generating a logic relationship corresponding to the text information; a drawing instruction is generated according to the logic relation, so that drawing software generates an attached drawing of the patent application according to the drawing instruction, and the drawing instruction can be recognized by the drawing software. According to the method, the working efficiency can be improved, the purpose of improving the efficiency of generating the attached drawings of the patent application is achieved, the time and the cost of making the attached drawings by a patent applicant can be reduced, and the technical problems of low efficiency and relatively high cost caused by manual completion of the attached drawings of the patent application in related technologies are solved.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and in particular relates to a method for generating drawings for a patent application, a device for generating drawings for a patent application, a computer-readable storage medium and an electronic device. Background Art

[0002] In the related art, when using artificial intelligence as an auxiliary tool to write patents, there are obvious limitations. It is mainly a single-modal model, that is, it can only be from text to text, that is, the input and output are both in text format. However, the patent field needs to reflect the technical solution of the patent application through drawings. The drawings of the patent application still need to be drawn by professionals, which takes a lot of time and cost, making it inefficient and costly. Summary of the invention

[0003] The present invention aims to solve one of the technical problems in the related art at least to a certain extent. To this end, the first object of the present invention is to propose a method for generating drawings of a patent application.

[0004] A second object of the present invention is to provide a device for generating drawings for a patent application.

[0005] A third object of the present invention is to provide a computer-readable storage medium.

[0006] A fourth objective of the present invention is to provide an electronic device.

[0007] To achieve the above-mentioned purpose, the first aspect of the present invention proposes a method for generating drawings of a patent application, the method is based on a large language model, and the method includes: obtaining text information of the patent application, the text information including the claims and / or description of the patent application; parsing the inherent logic in the text information, and generating a logical relationship corresponding to the text information; generating drawing instructions according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instructions, wherein the drawing instructions can be recognized by the drawing software.

[0008] According to the method for generating drawings of a patent application in an embodiment of the present invention, the claims and / or description of the patent application are first obtained, and then the internal logic in the claims and / or description of the patent application is parsed, and a logical relationship corresponding to the patent application is generated, and finally a drawing instruction is generated according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instruction. As a result, the method can improve work efficiency, achieve the purpose of improving the efficiency of generating drawings of patent applications, and reduce the time and cost of patent applicants to produce drawings, thereby solving the technical problem that the drawings of patent applications in the related art are completed manually, resulting in low efficiency and high cost.

[0009] In addition, the method for generating drawings of a patent application according to the above embodiment of the present invention may also have the following additional technical features: According to an embodiment of the present invention, at least one of the logical relationship, the drawing instructions and the drawings may be exported and edited.

[0010] According to one embodiment of the present invention, the method further includes: feeding back the edited logical relationship, the drawing instruction or the illustration to the large language model, so that the large language model is optimized according to the edited logical relationship, the drawing instruction or the illustration.

[0011] According to one embodiment of the present invention, the drawing software includes CAD, Solidworks, Visio, and AI drawing software.

[0012] According to one embodiment of the present invention, when the patent application involves a mechanical structure, the logical relationship is the connection relationship and / or relative position relationship between the components constituting the mechanical structure.

[0013] According to one embodiment of the present invention, when the patent application involves a circuit structure, the logical relationship is the connection relationship between the components constituting the circuit structure.

[0014] According to one embodiment of the present invention, when the patent application involves a method flow, the logical relationship is each step of the method flow that is sequentially connected.

[0015] To achieve the above-mentioned purpose, the second aspect of the present invention proposes a device for generating drawings of a patent application, the device is based on a large language model, and the device includes: an acquisition module, used to acquire text information of the patent application, the text information includes the claims and / or description of the patent application; a parsing module, used to parse the inherent logic in the text information and generate a logical relationship corresponding to the text information; a generation module, used to generate a drawing instruction according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instruction, wherein the drawing instruction can be recognized by the drawing software.

[0016] According to the device for generating drawings of a patent application in an embodiment of the present invention, the acquisition module acquires the claims and / or description of the patent application, the parsing module parses the internal logic in the claims and / or description of the patent application, and generates a logical relationship corresponding to the patent application, and the generation module generates a drawing instruction according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instruction. Thus, the device can improve work efficiency, achieve the purpose of improving the efficiency of generating drawings of patent applications, and reduce the time and cost of patent applicants to make drawings, thereby solving the technical problem that the drawings of patent applications in the related art are completed manually, resulting in low efficiency and high cost.

[0017] To achieve the above objectives, the third aspect of the present invention proposes a computer-readable storage medium on which a program for generating drawings for a patent application is stored. When the program for generating drawings for a patent application is executed by a processor, the method for generating drawings for a patent application according to the above-mentioned embodiment is implemented.

[0018] According to the computer-readable storage medium of an embodiment of the present invention, by executing the above-mentioned method for generating drawings for patent applications, it is possible to improve work efficiency, achieve the purpose of improving the efficiency of generating drawings for patent applications, and reduce the time and cost of patent applicants in making drawings, thereby solving the technical problem in the related art that drawings for patent applications are manually completed, resulting in low efficiency and high cost.

[0019] To achieve the above-mentioned purpose, the fourth aspect of the present invention proposes an electronic device, comprising: a memory and a processor, wherein the memory stores a program that can be run on the processor, and when the processor executes the program, it implements the method of generating drawings of a patent application according to the above-mentioned implementation method.

[0020] According to the electronic device of the embodiment of the present invention, by executing the above-mentioned method for generating drawings for patent applications, it is possible to improve work efficiency, achieve the purpose of improving the efficiency of generating drawings for patent applications, and reduce the time and cost of patent applicants in making drawings, thereby solving the technical problem in the related art that drawings for patent applications are manually completed, resulting in low efficiency and high cost.

[0021] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A flowchart of a method for generating drawings for a patent application according to an embodiment of the present invention; Figure 2 A block diagram of an apparatus for generating drawings for a patent application according to an embodiment of the present invention; Figure 3 FIG. 4 is a block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0024] The following describes a method for generating drawings for a patent application, an apparatus for generating drawings for a patent application, a computer-readable storage medium, and an electronic device according to an embodiment of the present invention with reference to the accompanying drawings.

[0025] Figure 1 A flowchart of a method for generating drawings for a patent application according to an embodiment of the present invention.

[0026] like Figure 1 As shown, the method for generating drawings of a patent application according to an embodiment of the present invention is based on a large language model and may include the following steps: S1, obtaining text information of the patent application, the text information including the claims and / or description of the patent application.

[0027] S2, parses the internal logic in the text information and generates a logical relationship corresponding to the text information.

[0028] S3, generating drawing instructions according to the logical relationship, so that the drawing software generates drawings of the patent application according to the drawing instructions, wherein the drawing instructions can be recognized by the drawing software.

[0029] Specifically, after a user (such as a patent applicant) completes writing the text portion of a patent application using a terminal device (such as a computer), the user can input the claims and / or the specification of the patent application into a large language model. It should be understood that the large language model can be a pre-trained model, and the excellent patent application documents related to various fields can be used to input into the large language model, so that the large language model generates corresponding drawings, and the generated drawings are compared with the drawings of the excellent patent application documents, and the large language model is corrected according to the difference between the two, thereby completing the pre-training of the large language model. Furthermore, since different patent applications involve different information, the embodiment of the present invention does not limit the specific form of the large language model, as long as it can realize the analysis of text information. In one embodiment of the present invention, the large language model can be a BERT (Bidirectional Encoder Representations from Transformers) model, that is, a bidirectional encoder representation transformer model.

[0030] Specifically, after obtaining the text information of the patent application, the large language model can first divide the text information into multiple sentences, and then perform semantic understanding on each sentence to determine the meaning expressed by each sentence. Next, multiple sentences with the same meaning can be merged together to form a paragraph, and then the paragraph can be semantically understood to determine the theme to be expressed in the paragraph. Finally, multiple paragraphs under the same topic can be merged to form a complete text information. By performing a step-by-step semantic understanding of the text information, the internal logic in the text information can be finally obtained, and the logical relationship text corresponding to the text information can be generated and saved.

[0031] Furthermore, the large language model can automatically generate drawing instructions based on logical relationships, wherein the drawing instructions can be files in SVG format, whereby the drawing instructions can be directly recognized by the drawing software, so that the drawing software can directly generate drawings based on the drawing instructions; alternatively, the large language model can also first convert the logical relationship into a file in an intermediate format, and then convert the file in the intermediate format into a drawing instruction. For example, the large language model can convert the logical relationship into a file in JSON format, and then convert the file in JSON format into a file in SVG format. The large language model sends the drawing instructions to the drawing software, and the drawing software can generate drawings of the patent application based on the drawing instructions for the user to use. It should be noted that the drawing software can be any drawing software that supports the SVG format, such as CAD, Solidworks, Visio, AI drawing software, and the like.

[0032] Therefore, the method for generating drawings for a patent application in an embodiment of the present invention obtains text information of the patent application, then parses the internal logic in the text information, generates a logical relationship corresponding to the text information, and finally generates a drawing instruction according to the logical relationship, so that the drawing software generates the drawings for the patent application according to the drawing instruction, thereby improving work efficiency, achieving the purpose of improving the efficiency of generating drawings for patent applications, and reducing the time and cost for applicants to prepare drawings, thereby solving the technical problem in the prior art that the drawings for patent applications are completed manually, resulting in low efficiency and high cost.

[0033] According to one embodiment of the present invention, at least one of a logical relationship, a drawing instruction, and a drawing figure may be exported and edited.

[0034] That is to say, drawings, logical relationships and drawing instructions can also be exported for users to view and edit. Specifically, users can edit the parsed logical relationships to correct unreasonable logical relationships. For example, users can check whether the exported logical relationships correspond to the internal logic of the patent application. If not, the logical relationships can be edited and modified, so that the large language model can generate drawing instructions based on the modified logical relationships to obtain accurate drawings. Users can also modify the generated drawing commands to optimize the drawing process or correct errors. Users can also edit the generated drawings, such as adjusting the layout, modifying the size or updating the style. As a result, users can not only modify the drawings themselves, but also modify the logical relationships and drawing instructions to ensure that the resulting drawings are more accurate.

[0035] According to an embodiment of the present invention, the method further includes: feeding back the edited logical relationship, drawing instruction or illustration to the large language model, so that the large language model is optimized according to the edited logical relationship, drawing instruction or illustration.

[0036] Specifically, after receiving the edited logical relationship, drawing instructions or drawings, the large language model analyzes the collected feedback to determine in which aspects the large language model needs to be improved. For example, if the feedback indicates that certain connection relationships are not correctly identified, then the relevant parts of the large language model need to be adjusted. The large language model is then adjusted by modifying the training data of the large language model, adjusting the algorithm parameters or improving the natural language processing capabilities of the large language model, and the adjusted large language model is used for retraining so that the large language model can learn the new logical relationship and the generation method of drawing instructions.

[0037] Furthermore, after the large language model is retrained, the performance of the new and old large language models on the same data set is compared for verification and testing to evaluate whether its performance has improved. If the test results show that the performance of the large language model has improved, the new large language model can be used for actual patent application drawing generation. If the performance does not meet expectations, further analysis and adjustment may be required, and then the above process is repeated. Finally, over time, new feedback is continuously collected and used for continuous optimization of the large language model, thereby ensuring that the large language model is always in the best state and can accurately understand and generate complex logical relationships and drawing instructions. Therefore, by continuously optimizing the large language model based on feedback, the accuracy of the large language model can be continuously improved, thereby improving the adaptability and accuracy of the large language model.

[0038] According to one embodiment of the present invention, when a patent application involves a mechanical structure, the logical relationship is the connection relationship and / or relative position relationship between the components constituting the mechanical structure.

[0039] Specifically, when a patent application involves a mechanical structure, the claims and / or the specification usually record the connection relationship and / or relative position relationship between the various components that make up the mechanical structure. For example, the connection relationship between the various components includes: rigid connection, hinged connection, sliding connection, elastic connection, etc., and the relative position relationship between the various components includes: parallel, vertical, specific angle, symmetry, etc. The large language model parses the claims and / or specifications of the patent application, and the obtained logical relationship includes the connection relationship and / or relative position relationship between the various components, so that the generated drawings are more accurate. It should be understood that when the drawing software executes the drawing instructions, it can call the graphic modules stored in the graphics library as the components of the mechanical structure, and can add labels to each component according to the figure marks recorded in the specification.

[0040] For example, the content of a patent application for a mechanical structure is as follows: "An improved gear transmission system includes a driving gear 10, a driven gear 20 and a fixed frame 30. The driving gear 10 is fixed to the driving shaft 40 through a key connection 11. The driven gear 20 is mounted on the fixed frame 30 through a bearing 21 and meshes with the driving gear 10. The fixed frame 30 is designed to be parallel to the machine base 50 and is fixed to the machine base by bolts 31." Then the logical relationships of the application parsed by the large language model include: Connection relationship: the driving gear 10 and the driving shaft 40 are rigidly connected, which is achieved through the key connection 11; the driven gear 20 and the fixing frame 30 are hingedly connected, which is achieved through the bearing 21; the fixing frame 30 and the base 50 are rigidly connected, which is achieved through the bolt 31; Relative position relationship: the driving gear 10 and the driven gear 20 are in the same vertical plane, and the axes are parallel; the fixing frame 30 is parallel to the base 50, and the fixing frame is located directly above the base.

[0041] The large language model can generate corresponding drawing instructions based on the logical relationship, so that the drawing software can generate drawings with component numbers. For complex structures, multiple views such as front view, side view, and cross-sectional view can also be generated to fully display the relationship between components.

[0042] According to one embodiment of the present invention, when a patent application involves a circuit structure, the logical relationship is the connection relationship between the components that make up the circuit structure.

[0043] Specifically, when a patent application involves a circuit structure, the logical relationship usually refers to the connection relationship between the various components in the circuit, including the series and parallel connection of the components and the signal flow between them.

[0044] For example, the content of a patent application for a circuit structure is as follows: "A high-efficiency audio amplifier includes: a power supply unit PS, an input buffer IB, a gain control unit GC, a power amplifier PA, an output buffer OB and a speaker SP, wherein the input buffer IB is used to receive an external audio signal; the gain control unit GC is used to adjust the gain of the external audio signal; the power amplifier PA is used to amplify the gain-adjusted audio signal; the output buffer OB is used to transmit the amplified audio signal to the speaker; wherein the input buffer IB, the gain control unit GC, the power amplifier PA and the output buffer OB are connected in sequence, and the power supply unit PS supplies power to the input buffer IB, the gain control unit GC, the power amplifier PA and the output buffer OB respectively". Then the logical relationship of the patent application analyzed by the large language model includes: "The power supply unit PS provides the necessary power for the entire audio amplifier; the input buffer IB receives the external audio signal and provides isolation to prevent signal interference; the gain control unit GC adjusts the gain of the audio signal to adapt to different input signal strengths; the power amplifier PA receives the signal output by the gain control unit GC and amplifies it to a power level sufficient to drive the speaker SP; the output buffer OB serves as an interface between the power amplifier PA and the speaker SP to ensure clean transmission of the signal; the speaker SP converts the electrical signal into sound." The large language model can generate corresponding drawing instructions based on the logical relationship to enable the drawing software to draw a circuit diagram. When drawing the circuit diagram, the drawing software (such as Visio) determines the layout and connection method of the components according to the drawing instructions, so that the drawing software generates a circuit diagram with component labels, wherein the circuit diagram will show the following connection relationship: the power supply unit PS is connected to the input buffer IB, the gain control unit GC, the power amplifier PA and the output buffer OB through a power line; the output end of the input buffer IB is connected to the input end of the gain control unit GC through a signal line; the output end of the gain control unit GC is connected to the input end of the power amplifier PA through a signal line; the output end of the power amplifier PA is connected to the output buffer OB through a speaker line, and then connected to the speaker SP.

[0045] According to one embodiment of the present invention, when a patent application involves a method flow, the logical relationship is each step of the method flow connected sequentially.

[0046] Specifically, for patent applications for method flows, the logical relationship is reflected in the sequential connection of the steps in the flow, and the steps are usually executed in a specific order to achieve a specific method or process.

[0047] For example, the content of a patent application for a method flow is as follows: "An automated quality inspection method comprises the following steps: Step 1: Product transportation: transport the product to be tested to the testing area; Step 2: Image acquisition: Use a camera to capture images of the product; Step three: image processing, processing the collected images; Step 4: Feature identification, identifying product features and defects; Step 5: Data analysis: analyzing the identified features; Step 6: Output the results and classify the products according to the analysis results; Step seven, feedback control, feeds the results back to the production line. ” The logical relationships of the patent application parsed by the large language model include: sequentiality: the execution of each step depends on the completion of the previous step. For example, image acquisition can only be carried out after the product is transported to the inspection area. Connectivity: There is a logical connection between the steps. The output of one step is the input of the next step. For example, the output of image processing is a preprocessed image for feature recognition.

[0048] The large language model can generate corresponding drawing instructions based on the logical relationship to enable the drawing software to draw a flowchart, where the flowchart is presented in the form of a flow chart, where each box represents a step; arrows indicate the flow between steps to show the order from one step to the next; the flowchart may also include decision nodes, such as "yes / no" judgments, which are used to indicate classification decisions made based on data analysis results.

[0049] In summary, according to the method for generating drawings of a patent application in an embodiment of the present invention, the claims and / or description of the patent application are first obtained, and then the internal logic in the claims and / or description of the patent application is parsed, and a logical relationship corresponding to the patent application is generated, and finally a drawing instruction is generated according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instruction. Therefore, the method can improve work efficiency, achieve the purpose of improving the efficiency of generating drawings of patent applications, and reduce the time and cost of patent applicants to produce drawings, thereby solving the technical problem that the drawings of patent applications in the related art are completed manually, resulting in low efficiency and high cost.

[0050] Corresponding to the above-mentioned embodiment, the present invention also proposes a device for generating drawings of a patent application.

[0051] Figure 2 A block diagram of an apparatus for generating drawings for a patent application according to an embodiment of the present invention.

[0052] like Figure 2 As shown, the device 100 for generating drawings of a patent application according to an embodiment of the present invention is based on a large language model and may include: an acquisition module 110 , a parsing module 120 and a generation module 130 .

[0053] The acquisition module 110 is used to acquire the text information of the patent application, and the text information includes the claims and / or the specification of the patent application. The parsing module 120 is used to parse the internal logic in the text information and generate the logical relationship corresponding to the text information. The generation module 130 is used to generate drawing instructions according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instructions, wherein the drawing instructions can be recognized by the drawing software.

[0054] According to one embodiment of the present invention, at least one of a logical relationship, a drawing instruction, and a drawing figure may be exported and edited.

[0055] According to one embodiment of the present invention, the above-mentioned device 100 also includes a feedback module, which is used to feed back the edited logical relationship, drawing instruction or illustration to the large language model, so that the large language model is optimized according to the edited logical relationship, drawing instruction or illustration.

[0056] According to one embodiment of the present invention, the drawing software includes CAD, Solidworks, Visio, and AI drawing software.

[0057] According to one embodiment of the present invention, when a patent application involves a mechanical structure, the logical relationship is the connection relationship and / or relative position relationship between the components constituting the mechanical structure.

[0058] According to one embodiment of the present invention, when a patent application involves a circuit structure, the logical relationship is the connection relationship between the components that make up the circuit structure.

[0059] According to one embodiment of the present invention, when a patent application involves a method flow, the logical relationship is each step of the method flow connected sequentially.

[0060] It should be noted that for details not disclosed in the device for generating drawings of a patent application in an embodiment of the present invention, please refer to the details disclosed in the method for generating drawings of a patent application in an embodiment of the present invention, and the details will not be repeated here.

[0061] According to the device for generating drawings of a patent application in an embodiment of the present invention, the acquisition module acquires the claims and / or description of the patent application, the parsing module parses the internal logic in the claims and / or description of the patent application, and generates a logical relationship corresponding to the patent application, and the generation module generates a drawing instruction according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instruction. Thus, the device can improve work efficiency, achieve the purpose of improving the efficiency of generating drawings of patent applications, and reduce the time and cost of patent applicants to make drawings, thereby solving the technical problem that the drawings of patent applications in the related art are completed manually, resulting in low efficiency and high cost.

[0062] Corresponding to the above embodiment, the present invention also proposes a computer-readable storage medium.

[0063] The present invention also provides a computer-readable storage medium on which is stored a program for generating drawings for a patent application. When the program for generating drawings for a patent application is executed by a processor, the method for generating drawings for a patent application according to the above-mentioned embodiment is implemented.

[0064] According to the computer-readable storage medium of an embodiment of the present invention, by executing the above-mentioned method for generating drawings for patent applications, it is possible to improve work efficiency, achieve the purpose of improving the efficiency of generating drawings for patent applications, and reduce the time and cost of patent applicants in making drawings, thereby solving the technical problem in the related art that drawings for patent applications are manually completed, resulting in low efficiency and high cost.

[0065] Corresponding to the above embodiment, the present invention further provides an electronic device.

[0066] Figure 3 FIG. 4 is a block diagram of an electronic device according to an embodiment of the present invention.

[0067] like Figure 3 As shown, the electronic device 200 of an embodiment of the present invention includes: a memory 210 and a processor 220. The memory 210 stores a program that can be run on the processor 220. When the processor 220 executes the program, the method of generating drawings of the patent application according to the above-mentioned embodiment is implemented.

[0068] According to the electronic device of the embodiment of the present invention, by executing the above-mentioned method for generating drawings for patent applications, it is possible to improve work efficiency, achieve the purpose of improving the efficiency of generating drawings for patent applications, and reduce the time and cost of patent applicants in making drawings, thereby solving the technical problem in the related art that drawings for patent applications are manually completed, resulting in low efficiency and high cost.

[0069] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0070] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0071] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0072] In addition, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0073] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0074] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for generating drawings for a patent application, characterized in that The method is based on a large language model and comprises: Obtaining text information of the patent application, wherein the text information includes claims and / or description of the patent application; Analyzing the inherent logic in the text information and generating a logical relationship corresponding to the text information; A drawing instruction is generated according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instruction, wherein the drawing instruction can be recognized by the drawing software.

2. The method for generating drawings for a patent application according to claim 1, characterized in that: At least one of the logical relationship, the drawing instructions, and the drawings may be exported and edited.

3. The method for generating drawings for a patent application according to claim 2, characterized in that: The method further includes: feeding back the edited logical relationship, the drawing instruction or the drawing to the large language model, so that the large language model is optimized according to the edited logical relationship, the drawing instruction or the drawing.

4. The method for generating drawings for a patent application according to claim 1, characterized in that: The drawing software includes CAD, Solidworks, Visio, and AI drawing software.

5. The method for generating drawings for a patent application according to any one of claims 1 to 4, characterized in that: When the patent application involves a mechanical structure, the logical relationship is the connection relationship and / or relative position relationship between the components constituting the mechanical structure.

6. The method for generating drawings for a patent application according to any one of claims 1 to 4, characterized in that: When the patent application involves a circuit structure, the logical relationship is the connection relationship between the components that make up the circuit structure.

7. The method for generating drawings for a patent application according to any one of claims 1 to 4, characterized in that: When the patent application involves a method flow, the logical relationship is each step of the method flow that is connected sequentially.

8. A device for generating drawings for a patent application, characterized in that: The device is based on a large language model, and comprises: An acquisition module, used to acquire text information of the patent application, wherein the text information includes claims and / or description of the patent application; A parsing module, used to parse the internal logic in the text information and generate a logical relationship corresponding to the text information; A generation module is used to generate a drawing instruction according to the logical relationship, so that the drawing software generates the drawings of the patent application according to the drawing instruction, wherein the drawing instruction can be recognized by the drawing software.

9. A computer-readable storage medium, characterized in that: A program for generating drawings for a patent application is stored thereon. When the program for generating drawings for a patent application is executed by a processor, a method for generating drawings for a patent application according to any one of claims 1-7 is implemented.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a program that can be run on the processor, and when the processor executes the program, the method for generating drawings of a patent application according to any one of claims 1 to 7 is implemented.

Images