Information processing device, information processing method, and information processing program
The device and method improve generative AI's understanding of patent claims by generating and presenting instruction statements to a trained model to extract and correlate unclear claim sections with technical documents, enhancing accuracy.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- TOYOTA JIDOSHA KK
- Filing Date
- 2024-12-26
- Publication Date
- 2026-07-08
AI Technical Summary
Generative AI struggles to accurately understand the content of patent claims due to abstracted descriptions, leading to incorrect interpretations.
An information processing device and method that generates and presents instruction statements to a generation model to extract and correlate parts of patent claims that are not clearly defined, using a trained model like GPT-3 or BERT, to identify corresponding sections in technical documents.
Enhances the accuracy of generative AI's understanding of patent claims by clearly identifying and correlating abstracted sections with relevant technical information, reducing errors in interpretation.
Smart Images
Figure 2026114625000001_ABST
Abstract
Description
Technical Field
[0004] , , ,
[0005] , , ,
[0001] The present disclosure relates to an information processing apparatus, an information processing method, and an information processing program
Background Art
[0002] There are techniques for assisting in the understanding of patent documents. Patent Document 1 discloses a technique aimed at enabling the focused reading of the description of the constituent elements that are key points in each claim of the claims in a patent document by identifying and presenting the constituent elements that are key points in each claim of the claims in the patent document.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] Consider a case where a generative AI is made to understand the description of each claim in the claims of a patent and extract where the content of each claim corresponds in another technical document. The description of each claim in the claims of a patent does not always represent the description in the detailed description of the invention as it is, and there are quite a few that are abstracted such as "first ~" and "second ~". The description of a claim that does not represent the description in the detailed description of the invention as it is hinders the understanding of the description content of the claim by the generative AI and may also lead to an incorrect understanding of the description content of the claim. <00,00027> The present disclosure has been made in view of the above points, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program capable of accurately making a generative AI understand the description content of a claim even when the description of each claim in the claims of a patent is abstracted.
Means for Solving the Problems
[0006] An information processing device according to a first aspect of this disclosure includes a generation unit that generates instruction statements for instructing a generation model, and a presentation unit that presents the output from the generation model, wherein the generation unit generates a first instruction statement instructing the generation model to extract parts of a document relating to the scope of patent rights where the technology cannot be specified, generates a second instruction statement instructing the generation model to extract correspondences between the parts extracted by the generation model based on the first instruction statement and technical documents related to the document relating to the scope of patent rights, and the presentation unit presents the correspondences extracted by the generation model based on the second instruction statement.
[0007] An information processing device according to a second aspect of this disclosure is an information processing device according to a first aspect, wherein the technical document is a specification of an application containing a document relating to the scope of the patent rights or a standard document relating to a document relating to the scope of the patent rights.
[0008] The information processing device relating to the third aspect of this disclosure is the information processing device relating to the first aspect, and the text relating to the scope of the patent is the text of the claims of the patent document.
[0009] An information processing method according to a fourth aspect of this disclosure involves a processor generating a first instruction statement that instructs a generation model to extract portions of a document relating to the scope of patent rights where the technology cannot be identified; generating a second instruction statement that instructs the generation model to extract correspondences between the portions extracted by the generation model based on the first instruction statement and technical documents related to the document relating to the scope of patent rights; and performing a process of presenting the correspondences extracted by the generation model based on the second instruction statement.
[0010] An information processing program according to a fifth aspect of this disclosure causes a computer to generate a first instruction statement that instructs a generation model to extract portions of a document relating to the scope of patent rights where the technology cannot be identified; generate a second instruction statement that instructs the generation model to extract correspondences between the portions extracted by the generation model based on the first instruction statement and technical documents related to the document relating to the scope of patent rights; and execute a process that presents the correspondences extracted by the generation model based on the second instruction statement. [Effects of the Invention]
[0011] According to this disclosure, it is possible to provide an information processing device, an information processing method, and an information processing program that enable a generating AI to accurately understand the content of each claim in the patent claims, even when the description of each claim is abstracted. [Brief explanation of the drawing]
[0012] [Figure 1] This figure shows an overview of an information processing device according to an embodiment of the disclosed technology. [Figure 2] This is a block diagram showing the hardware configuration of an information processing device. [Figure 3] This is a block diagram showing an example of the functional configuration of an information processing device. [Figure 4] This figure shows an example of a GUI being presented to the display unit by the presentation unit. [Figure 5] This figure shows an example of a GUI being presented to the display unit by the presentation unit. [Figure 6] This is a flowchart showing the flow of information processing by an information processing device. [Modes for carrying out the invention]
[0013] Hereinafter, an example of an embodiment of this disclosure will be described with reference to the drawings. In each drawing, identical or equivalent components and parts are given the same reference numerals. Also, the dimensional ratios in the drawings are exaggerated for illustrative purposes and may differ from the actual ratios.
[0014] Figure 1 is a diagram showing an overview of the information processing apparatus according to the present embodiment. The information processing apparatus 10 according to the present embodiment is an apparatus for assisting in the understanding of patent documents. The patent document refers to a document related to a patent such as a patent publication bulletin or a patent gazette. A patent document generally includes claims, a detailed description of the invention, an abstract, and drawings. The information processing apparatus 10 is an apparatus that causes the generation model 20 to read a patent document and a technical document to be compared with the patent document, gives a predetermined instruction sentence (prompt) to the generation model 20, and causes the generation model 20 to output the correspondence between the description of the claims of the patent document and the technical document. Here, the technical document includes a patent document, a technical specification, a technical standard, a product manual, a product pamphlet, and the like.
[0015] The generation model 20 is a trained model that has been pre-trained to generate and output text, images, etc. in response to an input. For example, as a model that generates text in response to an input, the generation model 20 may use GPT series (e.g., GPT-3, GPT-), BERT, RoBERTa, etc.
[0016] Note that the generation model 20 may be stored in a device different from the information processing apparatus 10, or may be stored in the same device as the information processing apparatus 10.
[0017] Figure 2 is a block diagram showing the hardware configuration of the information processing apparatus 10.
[0018] As shown in Figure 2, the information processing apparatus 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and a communication interface (I / F) 17. Each component is connected to be communicable with each other via a bus 19.
[0019] The CPU 11 is a central processing unit that executes various programs and controls each part. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a working area. The CPU 11 performs control of each of the above configurations and various arithmetic processes according to the program recorded in the ROM 12 or the storage 14. In the present embodiment, an information processing program for causing the generation model 20 to understand the contents of the claims of the patent document is stored in the ROM 12 or the storage 14.
[0020] The ROM 12 stores various programs and various data. The RAM 13 temporarily stores a program or data as a working area. The storage 14 is composed of a storage device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a flash memory, and stores various programs including an operating system and various data.
[0021] The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used to perform various inputs.
[0022] The display unit 16 is, for example, a liquid crystal display, and displays various information. The display unit 16 may adopt a touch panel method and function as the input unit 15.
[0023] The communication interface 17 is an interface for communicating with other devices. For example, standards such as Ethernet (registered trademark), FDDI, Wi-Fi (registered trademark), etc. are used.
[0024] When executing the above information processing program, the information processing apparatus 10 realizes various functions using the above hardware resources. The functional configuration realized by the information processing apparatus 10 will be described.
[0025] FIG. 3 is a block diagram showing an example of the functional configuration of the information processing apparatus 10.
[0026] As shown in Figure 3, the information processing device 10 has a functional configuration consisting of a generation unit 101, an extraction unit 102, and a presentation unit 103. Each functional configuration is realized by the CPU 11 reading and executing an information processing program stored in the ROM 12 or storage 14.
[0027] The generation unit 101 generates instruction statements (prompts) for the generation model 20. The generation unit 101 starts generating instruction statements, for example, based on instructions from the user of the information processing device 10. Specifically, the generation unit 101 generates instruction statements that instruct the generation model 20 to extract parts of the claims of the patent document that do not specify the technology. The claims of the patent document are an example of text relating to the scope of rights of the patent in this disclosure. The instruction statement that instructs the generation model 20 to extract parts of the claims of the patent document that do not specify the technology is an example of the first instruction statement in this disclosure.
[0028] For example, the generation unit 101 generates an instruction statement that reads, "Please extract the parts of the following claims that do not specify the technology." The instruction statement could also read something like, "Please extract the parts that are not defined," or "Please extract the parts that are not interpreted in the claims." The generation unit 101 may further add specific details to the instruction statement. For example, the generation unit 101 may include in the instruction statement what parts do not specify the technology. Examples of parts that do not specify the technology include specific items with ordinal numbers such as "the first," "the second," etc.
[0029] Furthermore, the generation unit 101 generates an instruction statement that instructs the generation model 20 to extract the correspondence between the claims of the patent document and the technical documents related thereto, for any parts of the claims of the patent document extracted by the generation model 20 where the technology cannot be identified. The instruction statement that instructs the generation model 20 to extract the correspondence between the claims and the technical documents related thereto is an example of the second instruction statement of this disclosure.
[0030] For example, the generation unit 101 generates an instruction statement that reads the technical document related to the claims and extracts the configuration in the technical document that corresponds to the configuration of the claims below from the entire text of the technical document.
[0031] The extraction unit 102 provides the generation model 20 with the instruction statement generated by the generation unit 101 and obtains the text generated by the generation model 20. More specifically, the extraction unit 102 provides the generation model 20 with an instruction statement instructing it to extract the parts of the patent claim in the patent document that do not specify the technology, and obtains the text generated by the generation model 20, thereby extracting the parts of the patent claim in the patent document that do not specify the technology. Furthermore, the extraction unit 102 provides the generation model 20 with an instruction statement instructing it to extract the correspondence between the extracted parts of the patent claim in the patent claim and the technical documents related to the patent claim, and obtains the text generated by the generation model 20, thereby extracting the correspondence between the patent claim and the technical documents related to the patent claim.
[0032] The presentation unit 103 presents a graphical user interface (GUI) for the analysis of the claims of the patent document to the display unit 16. Specifically, the presentation unit 103 presents to the display unit 16 the results extracted by the generation model 20 by the extraction unit 102.
[0033] Figures 4 and 5 show examples of GUI presentations to the display unit 16 by the presentation unit 103. Figure 4 shows an example of presentations to the display unit 16 by the presentation unit 103, based on the results of the generation model 20 extracting parts of the patent claims in the patent document where the technology cannot be identified. In Figure 4, an example is shown in which the generation model 20 extracted "first frame," "second frame," "first information," and "first information" as parts where the technology cannot be identified.
[0034] Figure 5 shows an example of how the display unit 103 presents to the display unit 16 the results of the generation model 20 extracting the correspondence between the claims and the technical documents related to those claims, in cases where the technology in the claims of a patent document cannot be identified. Figure 5 shows an example in which the generation model 20 has extracted which parts of the technical documents related to the claims correspond to the "first frame," "second frame," and "first information."
[0035] With this configuration, the information processing device 10 can have the generation model 20 extract the correspondence between the patent claims and the technical documents related to those claims, even if the claims contain sections that cannot identify the technology for reasons such as being abstracted. With this configuration, the information processing device 10 can reduce errors in the generation model 20's understanding of the patent claims compared to when sections of the patent claims that cannot identify the technology are not extracted.
[0036] Next, the operation of the information processing device 10 will be explained.
[0037] Figure 6 is a flowchart showing the flow of information processing by the information processing device 10. Information processing is performed when the CPU 11 reads an information processing program from the ROM 12 or storage 14, loads it into the RAM 13, and executes it.
[0038] In step S101, the CPU 11 generates an instruction statement (the first instruction statement) that instructs the generation model 20 to extract parts of the claims of the patent document that do not specify the technology. For example, the CPU 11 generates an instruction statement that says, "Please extract parts of the following claims that do not specify the technology." The CPU 11 may also include in the instruction statement what parts do not specify the technology.
[0039] Following step S101, in step S102, the CPU 11 provides the claims of the patent document and the instruction statement (first instruction statement) generated in step S101 to the generation model 20, and obtains the text generated by the generation model 20 based on the instruction statement. By providing the claims of the patent document and the instruction statement generated in step S101 to the generation model 20, the CPU 11 allows the generation model 20 to extract parts of the claims of the patent document where the technology cannot be identified.
[0040] Following step S102, in step S103, the CPU 11 displays on the display unit 16 the portion of the patent claims in the patent document that was extracted by the generation model 20 in step S102 and in which the technology cannot be identified.
[0041] Following step S103, in step S104, the CPU 11 generates an instruction statement (second instruction statement) that instructs the generation model 20 to extract the correspondence between the claims of the patent document and the technical documents related to those claims, for the parts of the claims of the patent document that have been extracted by the generation model 20 that do not specify the technology. For example, the CPU 11 generates an instruction statement that says, "Read the technical documents related to the claims and extract the components in the technical documents that correspond to the following components of the claims from the entire text of the technical documents."
[0042] Following step S104, in step S105, the CPU 11 provides the generation model 20 with the technical document related to the claims of the patent document and the instruction statement (second instruction statement) generated in step S104, and retrieves the text generated by the generation model 20 based on the instruction statement. By providing the generation model 20 with the technical document related to the claims of the patent document and the instruction statement generated in step S104, the CPU 11 can extract the correspondence between the claims and the technical document related to the claims.
[0043] Following step S105, in step S106, the CPU 11 displays on the display unit 16 the correspondence between the claims of the patent document extracted by the generation model 20 in step S105 and the technical documents related to said claims.
[0044] By performing this process, the information processing device 10 enables the generation model 20 to extract the correspondence between the patent claims and the technical documents related to those claims, even if the descriptions of each claim in the patent claims contain sections that cannot identify the technology for reasons such as being abstracted. By performing this process, the information processing device 10 can reduce errors in the generation model 20's understanding of the patent claims compared to when sections of the patent claims in the patent documents that cannot identify the technology are not extracted.
[0045] While embodiments of the present disclosure have been described in detail above with reference to the attached drawings, the technical scope of the present disclosure is not limited to these examples. The embodiments described above are illustrative and do not limit the technical scope of the present disclosure. It is clear to any person with ordinary skill in the art of the present disclosure that various modifications or alterations may occur within the scope of the technical idea set forth in the claims, and these modifications or alterations will naturally also fall within the technical scope of the present disclosure.
[0046] Furthermore, the effects described in the above embodiments are descriptive or illustrative, and are not limited to those described in the above embodiments. In other words, the technology relating to this disclosure may produce other effects that would be obvious to a person of ordinary skill in the art of this disclosure from the descriptions in the above embodiments, in addition to or in lieu of the effects described in the above embodiments.
[0047] Furthermore, the information processing that the CPU reads and executes in each of the above embodiments may be executed by various processors other than the CPU. Examples of such processors include PLDs (Programmable Logic Devices) such as FPGAs (Field-Programmable Gate Arrays) whose circuit configuration can be changed after manufacturing, and dedicated electrical circuits that are processors with circuit configurations specifically designed to execute specific processing, such as ASICs (Application Specific Integrated Circuits). In addition, the information processing may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (for example, multiple FPGAs, and a combination of a CPU and an FPGA). More specifically, the hardware structure of these various processors is an electrical circuit that combines circuit elements such as semiconductor elements.
[0048] Furthermore, while the above embodiments describe a configuration in which the information processing program is pre-stored (installed) in ROM or storage, the invention is not limited thereto. The program may be provided in a form recorded on a non-transitory recording medium such as a CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Versatile Disk Read Only Memory), or USB (Universal Serial Bus) memory. The program may also be provided in a form that can be downloaded from an external device via a network. This disclosure may also be applied to program products. [Explanation of Symbols]
[0049] 10 Information Processing Devices 20 Generative Models
Claims
1. A generation unit that generates instruction statements to instruct the generative model, A display unit that presents the output from the aforementioned generation model, Equipped with, The generation unit generates a first instruction statement that instructs the generation model to extract portions of the text relating to the scope of patent rights in which the technology cannot be identified, and generates a second instruction statement that instructs the generation model to extract correspondences between the portions extracted by the generation model based on the first instruction statement and technical documents related to the text relating to the scope of patent rights. The presentation unit is an information processing device that presents the correspondence relationships extracted by the generation model based on the second instruction statement.
2. The information processing apparatus according to claim 1, wherein the technical document is the specification of an application containing a document relating to the scope of the patent rights or a standard document relating to a document relating to the scope of the patent rights.
3. The information processing apparatus according to claim 1, wherein the text relating to the scope of the patent is the text of the claims in the patent document.
4. The processor, A first instruction is generated that instructs the generative model to extract parts of the document concerning the scope of patent rights that do not specify the technology. Based on the first instruction, a second instruction is generated that instructs the generation model to extract the correspondence between the portion extracted by the generation model and the technical document related to the text concerning the scope of the patent rights. The correspondence extracted by the generative model based on the second instruction is presented. An information processing method that executes a process.
5. On the computer, A first instruction is generated that instructs the generative model to extract parts of the document concerning the scope of patent rights that do not specify the technology. Based on the first instruction, a second instruction is generated that instructs the generation model to extract the correspondence between the portion extracted by the generation model and the technical document related to the text concerning the scope of the patent rights. The correspondence extracted by the generative model based on the second instruction is presented. An information processing program that executes a process.