Information processing device, information processing method, and program

Abstract

To implement a business support technique applicable to various jobs.SOLUTION: An information processing device includes: specification means that specifies a target job category; generation means that generates a prompt according to the job category specified by the specification means; and acquisition means that obtains a generated sentence which is generated based on the prompt.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Technologies for supporting various tasks such as document creation and confirmation have been developed. For example, Patent Document 1 discloses a document creation program and an information processing device that generate legal documents based on the work of multiple people. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-166870 A Summary of the Invention [Problem to be solved by the invention]

[0004] Generally, tasks such as creating and checking documents are performed in various business operations. However, the technology described in Patent Document 1 has a problem in terms of applicability to various business operations.

[0005] The present disclosure has been made in consideration of the above problems, and has an object to provide a business support technology that can be applied to various businesses. [Means for solving the problem]

[0006] An information processing device according to one aspect of the present disclosure includes an identification means for identifying a target business category, a generation means for generating a prompt according to the business category identified by the identification means, and an acquisition means for acquiring a generated sentence generated based on the prompt.

[0007] An information processing method according to one aspect of the present disclosure includes at least one processor identifying a business category corresponding to a target sentence, generating a prompt according to the business category identified in the identification process, and obtaining the generated sentence based on the prompt.

[0008] A program relating to one aspect of the present disclosure causes a computer to execute a specification process that identifies a business category corresponding to a target sentence, a generation process that generates a prompt according to the business category identified in the specification process, and an acquisition process that acquires the generated sentence generated based on the prompt. Effect of the Invention

[0009] According to the present disclosure, it is possible to provide a business support technology that can be applied to various businesses. [Brief description of the drawings]

[0010] [Figure 1] 1 is a block diagram showing a configuration of an information processing device according to the present disclosure. [Diagram 2] 1 is a flow diagram showing the flow of an information processing method according to the present disclosure. [Diagram 3] 1 is a block diagram showing a configuration of an information processing system according to the present disclosure. [Figure 4] FIG. 11 is a flow diagram showing the flow of an element extraction process according to the present disclosure. [Diagram 5] FIG. 11 is a diagram for explaining an element extraction process according to the present disclosure. [Figure 6] 11 is a flow diagram showing the flow of a prompt generation and generated sentence acquisition process according to the present disclosure. [Figure 7] 11 is a diagram for explaining a process for generating a prompt and obtaining a generated sentence according to the present disclosure. FIG. [Figure 8] 11 is a diagram for explaining a process for generating a prompt and obtaining a generated sentence according to the present disclosure. FIG. [Figure 9] 11 is a diagram for explaining a process for generating a prompt and obtaining a generated sentence according to the present disclosure. FIG. [Figure 10]11 is a diagram for explaining a process for generating a prompt and obtaining a generated sentence according to the present disclosure. FIG. [Figure 11] 11 is a diagram for explaining a process for generating a prompt and obtaining a generated sentence according to the present disclosure. FIG. [Figure 12] 11 is a diagram for explaining a process for generating a prompt and obtaining a generated sentence according to the present disclosure. FIG. [Figure 13] 1 is a block diagram showing a hardware configuration of an information processing device according to the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] The following are examples of embodiments of the present invention. However, the present invention is not limited to the embodiments exemplified below, and various modifications are possible within the scope of the claims. For example, an embodiment obtained by appropriately combining the technical means employed in each of the embodiments exemplified below may also be included in the scope of the present invention. Alternatively, an embodiment obtained by appropriately omitting a part of the technical means employed in the embodiments exemplified below may also be included in the scope of the present invention. In addition, the effects mentioned in each of the exemplary embodiments shown below are examples of effects expected in the exemplary embodiments, and do not define the scope of the present invention. In other words, an embodiment that does not exhibit the effects mentioned in each of the exemplary embodiments shown below may also be included in the scope of the present invention.

[0012] [First embodiment] A first exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is the basic form of each exemplary embodiment described later. The scope of application of each technical means adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means adopted in this exemplary embodiment can be adopted in other exemplary embodiments included in this disclosure to the extent that no particular technical obstacle occurs. In addition, each technical means shown in the drawings referred to for explaining this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure to the extent that no particular technical obstacle occurs.

[0013] (Configuration of information processing device 1) The configuration of an information processing device 1 according to this exemplary embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the information processing device 1. As shown in Fig. 1, the information processing device 1 includes an identifying unit 11, a generating unit 12, and an acquiring unit 13.

[0014] (Specific part 11) The identification unit 11 identifies a target business category and provides category information indicating the identified business category to the generation unit 12, which will be described later. As an example, the identification unit 11 acquires an instruction from a user and identifies the target business category by referring to a wording included in the instruction. Alternatively, the identification unit 11 may be configured to acquire information indicating a user's selection regarding a business category and identify the target business category by referring to the information.

[0015] Here, the "business category" in this exemplary embodiment refers to a category related to at least one of an industry and a job type (job), for example. For example, the business category may include: ·Publishing industry ·Financial industry Other general industries In addition, each category may include, for example, Email Creation Title suggestions Needs survey · Estimate preparation · Contract drafting · Legal check of contracts Here, a certain subcategory may exist across multiple categories. For example, a subcategory called "email creation" may exist in any of the categories of the publishing industry, the finance industry, and other general industries. In this exemplary embodiment, the term "business category" may include the above-mentioned "industry category" and "job subcategory," but this is not intended to limit this exemplary embodiment.

[0016] (Generation unit 12) The generating unit 12 generates a prompt according to the task category identified by the identifying unit 11. Here, the prompt generated by the generating unit 12 is a prompt to be input to a language model described later, and the language model generates a generated sentence described later by referring to the content of the prompt. Note that the specific configuration of the prompt generated by the generating unit 12 does not limit this exemplary embodiment, but as an example, the generating unit 12 may be configured to execute an extraction process to extract one or more elements corresponding to the task category identified by the identifying unit 11 from one or more target sentences, and generate the prompt according to the elements extracted in the extraction process. Also, the generating unit 12 may be configured to generate the prompt including an instruction sentence corresponding to an instruction from a user and the element extracted in the extraction process. Note that the prompt generated by the generating unit 12 is a prompt including a natural sentence, but this does not limit this exemplary embodiment.

[0017] (Acquisition part 13) The acquisition unit 13 acquires the generated sentence generated based on the prompt generated by the generation unit 12. More specifically, the acquisition unit 13 inputs the prompt generated by the generation unit 12 into a language model, and acquires the sentence generated by the language model (also simply referred to as a "generated sentence"). Here, the language model may be a language model executed inside the information processing device 1, or a language model executed by a generation device communicably connected to the information processing device 1. Note that the language model is, as an example, a language model that refers to a prompt including a natural sentence and generates a generated sentence as a natural sentence, and is a language model that is machine-learned in advance, but this does not limit the present exemplary embodiment.

[0018] (Effects of information processing device 1) As described above, the information processing device 1 is configured to identify a target task category, generate a prompt corresponding to the identified task category, and obtain a sentence generated based on the prompt. Therefore, the information processing device 1 can provide a task support technology applicable to various tasks.

[0019] (Flow of information processing method S1) Next, the flow of the information processing method S1 according to this exemplary embodiment will be described with reference to Fig. 2. Fig. 2 is a flow diagram showing the flow of the information processing method S1. As shown in Fig. 2, the information processing method S1 includes a step (process) S11 of identifying a task category, a step (process) S12 of generating a prompt, and a step (process) S13 of acquiring a generated sentence.

[0020] (Step S11) In step S11, the identification unit 11 identifies a target business category, and provides category information indicating the identified business category to the generation unit 12, which will be described later. A more specific description of the identification unit 11 has been given above, and therefore will not be repeated here.

[0021] (Step S12) In step S12, the generator 12 generates a prompt according to the task category identified in step S11 by the identifier 11. The specific process of the generator 12 has been described above, and therefore will not be described here.

[0022] (Step S13) In step S13, the acquisition unit 13 acquires the generated sentence generated based on the prompt generated by the generation unit 12 in step S12. A more detailed description of the acquisition unit 13 has been given above, and therefore will not be repeated here.

[0023] (Effect of information processing method S1) As described above, the information processing method S1 is configured to identify a target task category, generate a prompt corresponding to the identified task category, and obtain a sentence generated based on the prompt. Therefore, the information processing method S1 can provide a task support technology that can be applied to various tasks.

[0024] Second Embodiment A second exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same functions as those described in the above exemplary embodiment will be given the same reference numerals, and their description will be omitted as appropriate. The scope of application of each technical means adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means adopted in this exemplary embodiment can be adopted in other exemplary embodiments included in this disclosure, as long as no particular technical hindrance occurs. In addition, each technical means shown in each drawing referred to for explaining this exemplary embodiment can be adopted in other exemplary embodiments included in this disclosure, as long as no particular technical hindrance occurs.

[0025] (Configuration of information processing system 1A) The configuration of an information processing system 1A according to this exemplary embodiment will be described with reference to Fig. 3. Fig. 3 is a block diagram showing the configuration of the information processing system 1A. As shown in Fig. 3, the information processing system 1A includes an information processing device 100 and a generating device 50 connected to the information processing device 100 via a network N. Here, the specific configuration of the network N does not limit this exemplary embodiment, but as an example, a wireless LAN (Local Area Network), a wired LAN, a WAN (Wide Area Network), a public line network, a mobile data communication network, or a combination of these networks can be used.

[0026] (Generation device 50) As shown in FIG. 3, the generation device 50 includes a control unit 51, a storage unit 52, and a communication unit 53. The communication unit 53 communicates with a device external to the generation device 50. As an example, the communication unit 53 communicates with an information processing device 100 included in the information processing system 1A. The communication unit 53 transmits data supplied from the control unit 51 to the information processing device 100, and supplies data received from the information processing device 100 to the control unit 51. Note that the data received by the communication unit 53 from the information processing device 100 may include a prompt generated by the information processing device 100. In addition, the data provided by the communication unit 53 to the information processing device 100 may include a generated sentence generated by a language model described later based on the prompt.

[0027] A language model LM is stored in the storage unit 52. As an example, a plurality of parameters that define the language model LM are stored in the storage unit 52. As an example, these parameters are parameters that have been learned in advance by machine learning (parameters that have undergone an update process by machine learning), but this does not limit the present exemplary embodiment.

[0028] The control unit 51 executes the language model to obtain an output result from the language model LM. As an example, the control unit 51 inputs a prompt received from the information processing device 100 to the language model LM, and obtains a generated sentence generated by the language model LM. The control unit 51 also provides the generated sentence to the information processing device 100 via the communication unit 53.

[0029] In the present exemplary embodiment, the generation device 50 is illustrated as a device separate from the information processing device 100, but this does not limit the present exemplary embodiment. The control unit 51 included in the generation device 50 or the function of the language model execution unit in the control unit 51 may be configured to be included in the control unit of the information processing device 100. Similarly, the language model LM stored in the storage unit 52 included in the generation device 50 may be stored in the storage unit of the information processing device 100, and the information processing device 100 itself may be configured to be able to execute the language model LM.

[0030] (Configuration of information processing device 100) Next, the configuration of the information processing device 100 according to this exemplary embodiment will be described with reference to Fig. 3. As shown in Fig. 3, the information processing device 100 includes a control unit 10, a storage unit 20, a communication unit 30, and an input / output unit 40.

[0031] (Communication unit 30) The communication unit 30 communicates with devices external to the information processing device 100. As an example, the communication unit 30 communicates with the generation device 50. The communication unit 53 transmits data supplied from the control unit 10 to the generation device 50, and supplies data received from the generation device 50 to the control unit 10. Note that the data transmitted by the communication unit 30 to the generation device 50 may include a prompt generated by the control unit 10. Furthermore, the data received by the communication unit 30 from the generation device 50 may include a generated sentence generated by the generation device 50 based on the prompt.

[0032] (Input / output section 40) The input / output unit 40 is configured to include at least one of an input / output device such as a keyboard, a mouse, a display, a printer, a touch panel, etc. Alternatively, the input / output unit 40 may be configured to have an input / output device such as a keyboard, a mouse, a display, a printer, a touch panel, etc. connected to it. In this configuration, the input / output unit 40 accepts various types of information input to the information processing device 100 from the connected input device. Moreover, the input / output unit 40 outputs various types of information to the connected output device under the control of the control unit 10. An example of the input / output unit 40 is an interface such as a Universal Serial Bus (USB).

[0033] (Storage unit 20) The storage unit 20 stores various data referenced by the control unit 10 and various data generated by the control unit 10. As an example, the storage unit 20 stores: The user instruction IN acquired by the acquisition unit 13 via the input / output unit 40 or the communication unit 30 ·One or more target sentences TS One or more elements EM extracted from the target sentence TS ·Multiple industry-specific models SM Prompt PR generated by the control unit 10 ·Generated sentence GS generated by the generator 50 - Modified generated sentence RGS The various data stored in the storage unit 20 will be specifically described later.

[0034] (Control unit 10) 3, the control unit 10 includes a determination unit 11, a generation unit 12, and an acquisition unit 13, similar to the first exemplary embodiment. In addition to these components, the control unit 10 includes a correction unit 14.

[0035] (Acquisition part 13) The acquisition unit 13 acquires a user instruction IN via the input / output unit 40 or the communication unit 30, and stores the acquired user instruction IN in the storage unit 20. For example, the user instruction IN may include -What kind of processing do you want to do in what industry category? The user instruction acquired by the acquisition unit 13 is referred to by the specification unit 11 and the generation unit 12, which will be described later.

[0036] The acquisition unit 13 also acquires the generated sentence GS generated by the generation device 50 based on the prompt PR generated by the generation unit 12 , and stores the generated sentence GS in the storage unit 20 .

[0037] (Specific part 11) As described in the first exemplary embodiment, the identification unit 11 identifies a target business category, and provides category information indicating the identified business category to the generation unit 12, which will be described later. As an example, the identification unit 11 identifies the target business category by referring to a wording included in the user instruction IN acquired by the acquisition unit 13. Alternatively, the identification unit 11 may be configured to acquire information indicating a user's selection included in the user instruction IN, and identify the target business category by referring to the information.

[0038] Here, the "business category" in this exemplary embodiment refers to a category related to at least one of industry and occupation (job), as an example, as described in the first exemplary embodiment.

[0039] (Generation unit 12) The generation unit 12 generates a prompt PR according to the task category identified by the identification unit 11 (prompt generation process). Here, the prompt PR generated by the generation unit 12 is a prompt to be input to the language model LM executed by the generation device 50. Note that the specific configuration of the prompt generated by the generation unit 12 is not limited to this exemplary embodiment, but as an example, the generation unit 12 may be configured to execute an extraction process (element extraction process) that extracts one or more elements EM corresponding to the task category identified by the identification unit 11 from one or more target sentences TS, and generate the prompt PR according to the element EM extracted in the extraction process.

[0040] Here, as an example, the generation unit 12 may use a language model (called an industry-specific model) SM trained for each business category in the extraction process. Here, the industry-specific model SM includes, as an example, A first industry-specific model SM1 trained by machine learning specialized for a first industry (first business category) A second industry-specific model SM2 trained by machine learning specialized for a second industry (second business category) A third industry-specific model SM3 trained by machine learning specialized for a third industry (third business category) More specifically, the industry-specific model SM includes: - Publishing industry email-specific model SM1 trained using training data containing multiple emails in the publishing industry - Financial industry contract-specific model SM2 trained using training data including multiple contracts in the financial industry - General industry quotation specialized model SM3 trained using training data including multiple quotations in other general industries The generating unit 12 may be configured to execute the extraction process by using an industry-specific model that is learned by specializing in the business category identified by the identifying unit 11. However, these examples do not limit the present exemplary embodiment.

[0041] The generation unit 12 may be configured to generate a prompt PR including a content corresponding to the user instruction IN. As an example, the generation unit 12 may be configured to generate the prompt PR including an instruction sentence corresponding to the user instruction IN and the element EM extracted in the extraction process. As an example, the prompt PR generated by the generation unit 12 is provided to the generation device 50 via the communication unit 30, and the generation device 50 uses the language model LM to generate a generated sentence GS based on the prompt PR. The generated sentence GS is acquired by the acquisition unit 13 described above via the communication unit 30. Specific examples of the prompt PR generated by the generation unit 12 and the generated sentence GS generated by the generation device 50 will be described later.

[0042] (revised part 14) The modification unit 14 modifies the generated sentences GS generated by the generation device 50. More specifically, the modification unit 14 modifies the generated sentences GS according to the business category identified by the identification unit 11. As an example, the modification unit 14 may be configured to modify the generated sentences GS using the above-mentioned industry-specific model SM. For example, the modification unit 14 may modify the generated sentences GS using an industry-specific model that has been trained to be specialized for the business category identified by the identification unit 11. The modified generated sentences RGS are provided to the user via the input / output unit 40 or the communication unit 30, as an example.

[0043] (Element extraction process by generation unit 12) Next, the flow of element extraction processing by the generation unit 12 will be described with reference to Fig. 4. As an example, the element extraction processing is performed prior to the prompt generation processing by the generation unit 12. In other words, as an example, the element extraction processing is performed as pre-processing of the prompt generation processing by the generation unit 12, but this is not intended to limit the present exemplary embodiment. Fig. 4 is a flow diagram showing the flow of element extraction processing by the generation unit 12.

[0044] (Step S111) In step S111, the acquisition unit 13 acquires one or more target sentences. Here, the acquisition unit 13 may be configured to acquire sentences belonging to a specific business category as the one or more target sentences, or may be configured to acquire sentences related to various business categories. The target sentences acquired in this step include: Email text ·Proposal Research Report · Quotation Contract The acquisition unit 13 may be configured to acquire templates of the above-mentioned various documents or parts of the templates as one or more target sentences.

[0045] (Step S112) In step S112, the generation unit 12 extracts one or more elements from one or more target sentences acquired in step S111. Here, the element may be any unit such as one phrase, multiple phrases, one clause, multiple clauses, one sentence, multiple sentences, one paragraph, multiple paragraphs, etc. The generation unit 12 may also extract the elements from the one or more target sentences using an industry-specific model trained to be specialized for the business category to which the one or more target sentences belong. As described above, the industry-specific model is a model trained to be specialized for each business category, so by using the industry-specific model, it is possible to accurately specify the extraction of elements.

[0046] Fig. 5 is a diagram for explaining an example of processing by the generation unit 12 in this step. In the example shown in Fig. 5, a contract is illustrated as a target sentence TS1, which is an example of a target sentence TS. The generation unit 12 extracts each clause included in the target sentence TS1 as elements EM1 to EM10, respectively.

[0047] (Step S113) In step S113, the generation unit 12 stores the elements extracted in step S112 in the storage unit 20. Here, as an example, the generation unit 12 stores each element in association with the business category to which the target sentence from which the element was extracted belongs. In the example shown in FIG. 5, the generation unit 12 stores the elements EM1 to EM10 in association with the business category of "contract" or "business of creating or editing a contract", which is the business category to which the target sentence from which the element was extracted belongs. However, these specific examples do not limit the present exemplary embodiment.

[0048] (Example 1 of prompt generation and generated sentence output processing) Next, a first example of a prompt generation and generated sentence output process executed by the information processing device 100 will be described with reference to Fig. 6. Fig. 6 is a flow diagram showing a flow of a first example of a prompt generation and generated sentence output process executed by the information processing device 100.

[0049] (Step S121) In step S121, the acquisition unit 13 acquires a user instruction IN via the input / output unit 40 or the communication unit 30, and stores the acquired user instruction IN in the storage unit 20. For example, the user instruction IN may include -What kind of processing do you want to do in what industry category? The information may include wording that suggests the following.

[0050] Fig. 7 is a diagram showing a user instruction IN1 which is an example of a user instruction acquired by the acquisition unit 13 in this step. In the example shown in Fig. 7, the user instruction IN1 includes a user instruction main text IN1_main and instruction elements INE1 to INE4 associated with the user instruction main text IN1_main.

[0051] (Step S11) In step S11, the identification unit 11 identifies the target business category by referring to the user instruction IN acquired in step S121. As an example, the identification unit 11 identifies the target business category as "contract" or "contract creation or editing work" by referring to one or more phrases included in the user instruction IN1 shown in Fig. 7.

[0052] As an example, the identification unit 11 may perform the category identification process using a language model provided in the generation device 50, or may perform the process using the various industry-specific models described above, or may perform the process using a combination of these models.

[0053] As an example, the specification unit 11 Generate a prompt for identifying an industry, the prompt including a user instruction IN1 and an instruction sentence for identifying a business category of the user instruction IN1; The generated industry-specific prompts are input into a language model included in the generation device 50; Language model: Refer to the output of the language model to identify the target business category. The following process may be performed.

[0054] As another example, the specification unit 11 Input user instructions IN1 to multiple industry-specific models, and identify the target business category by referring to the output from each industry-specific model. As described above, the industry-specific model is a model that is trained to be specialized for each business category, and therefore, by using the industry-specific model, it is possible to accurately identify the target business category.

[0055] Alternatively, the specification unit 11 may be configured to refer to a dictionary (keyword table) for each business category and specify the target business category depending on whether or not the user instruction IN contains a word in the dictionary.

[0056] In the example shown in Figure 7, the identification unit 11, as an example, refers to the words "contract" and "update" contained in the user instruction main text IN1_main, and "expiration date" contained in the instruction element INE1, and identifies that the target business category is "contract creation or editing work."

[0057] However, the business category identification process by the identification unit 11 is not limited to the above example, and the identification unit 11 may be configured to obtain information indicating the user's selection contained in the user instruction IN and identify the target business category by referring to the information.

[0058] (Step S122) Next, in step S122, the generation unit 12 acquires one or more elements related to the business category identified in step S11. As an example, the generation unit 12 acquires one or more elements related to the business category identified in step S11 from among the elements extracted in advance in the above-mentioned "element extraction process."

[0059] In this example, in step S11, the identification unit 11 identifies the target business category as "contract" or "contract creation or editing work," so in this step, the generation unit 12 acquires at least one of elements EM1 to EM10 (see Figure 5) associated with the business category as the one or more elements.

[0060] (Step S12) Next, in step S12, the generation unit 12 The user instruction acquired in step S121 The business category identified in step S11 One or more elements acquired in step S122 and generates a prompt PR according to at least one of the above.

[0061] 8 is a diagram showing a prompt PR1 which is an example of a prompt generated by the generation unit 12 in this step. As shown in FIG. 8, the prompt PR1 generated by the generation unit 12 includes the following: · Tags that indicate imperative statements (#imperative statements) · Command statement content (PR1_main) · Tags indicating constraints (#constraints) · Contents of constraints (INE1~INE4) · Tag indicating the input text (#input text) Input text content (EM6~EM8) · Tag indicating output statement (#output statement) Here, the content of the command statement (PR1_main) is the content generated by the generating unit 12 according to the business category identified by the identifying unit 11 in step S11. As an example, the generating unit 12 may generate the following: Manage multiple commands, each associated with a multiple business category, Selecting a command statement associated with the business category identified by the identification unit 11 from the plurality of command statements The command content (PR1_main) may be generated by performing the process shown in Fig. 8. In the example shown in Fig. 8, the command content (PR1_main) includes a command "You are a professional sales person. Please output the best edited result based on the following constraints and a part of the contract" corresponding to the target task category "contract creation or editing task."

[0062] On the other hand, the contents of the constraint conditions (INE1 to INE4) are contents generated by the generation unit 12 based on the information included in the user instruction IN acquired by the acquisition unit 13 in step S121. In the example shown in FIG. 8, the generation unit 12 uses the instruction elements INE1 to INE4 included in the user instruction IN1 shown in FIG. 7 as the contents of the constraint conditions. In other words, in the example shown in FIG. 8, the constraint conditions (INE1 to INE4) include the condition corresponding to the user instruction IN1. -Change the expiration date to two years. The validity period can be extended if both parties agree to it. · Use contractual language. Don’t leave out any important keywords. Contains:

[0063] The contents of the input sentence (EM6 to EM8) include one or more elements acquired by the generation unit 12 in step S122. In the example shown in FIG. 8, the elements EM6 to EM8 are included among the elements EM1 to EM10 (see FIG. 5) associated with the business category "contract creation or editing business". In the example shown in FIG. 8, as an example, the generation unit 12 extracts elements EM6 to EM8 that are relatively highly related to the user instruction IN1 from the elements EM1 to EM10, and uses them as the contents of the input sentence. Here, at least one of the above-mentioned language model LM and the industry-specific model may be used for the extraction process. However, these specific processes do not limit this exemplary embodiment, and the generation unit 12 may be configured to use all elements associated with the target business category as the contents of the input sentence.

[0064] In this way, the prompt PR generated by the generation unit 12 is, for example, A prompt containing content in response to a user instruction IN, The instruction sentence corresponding to the user instruction IN and the element EM extracted in the extraction process executed in advance in step S112 are included.

[0065] In this step, the generation unit 12 adds the above-mentioned · Command statement content (PR1_main) · Contents of constraints (INE1~INE4) Input text content (EM6~EM8) As an example, the generation unit 12 may generate the prompt PR by applying the following: · Tags that indicate imperative statements (#imperative statements) · Tags indicating constraints (#constraints) · Tag indicating the input text (#input text) The prompt PR may be generated by applying the above-mentioned command statement content (PR1_main), constraint condition content (INE1 to INE4), and input statement content (EM6 to EM8) to a prompt template including the above.

[0066] As another example, the generation unit 12 · Tags that indicate imperative statements (#imperative statements) · Tags indicating constraints (#constraints) · Tag indicating the input text (#input text) In addition, · Command statement content (PR1_main) The prompt PR may be generated by applying the above-mentioned constraint condition contents (INE1 to INE4) and the input sentence contents (EM6 to EM8) to a prompt template including the above.

[0067] The prompt PR generated in this step is provided to the generation device 50, for example, via the communication unit 30. Then, in the generation device 50, the prompt is input to the language model LM, and the language model LM generates a generated sentence based on the prompt.

[0068] The control unit 10 may be configured to present the generated prompt PR to the user via the input / output unit 40, or may not be configured to present the prompt PR.

[0069] (Step S13) Next, in step S13, the acquisition unit 13 acquires the generated sentence GS generated by the language model LM based on the prompt PR generated in step S12. The upper part of Fig. 9 shows a generated sentence GS1, which is an example of the generated sentence GS acquired by the acquisition unit 13 in this step. Note that in the example in the upper part of Fig. 9, the bold type and underline are mainly for the convenience of explanation in this specification, but as will be described later, they may also be used as decoration of the generated sentence GS when it is presented to the user.

[0070] 9, generated sentence GS1 includes elements GSE6 to GSE8. Here, in this example, element GSE6 is the same as element EM6 included in prompt PR1, and element GSE8 is the same as element EM8 included in prompt PR1.

[0071] On the other hand, the element GSE7 is modified by the language model LM with respect to the element EM7 included in the prompt PR1. More specifically, the constraint INE1 included in the prompt PR1 - Change the expiration date to 2 years In response to this, the expiration date has been changed to "2 years" by the language model LM.

[0072] In addition, the constraint INE2 included in prompt PR1 The validity period can be extended if both parties agree to it. In response to this, the following sentence has been added: "The validity period may be extended if both parties to the contract agree to it."

[0073] (Step S123) Next, in step S123, the correction unit 14 corrects the generated sentence GS acquired in step S13. As an example, the correction unit 14 corrects the generated sentence GS using an industry-specific model specialized for the target business category.

[0074] The lower part of Fig. 9 shows RGS1, which is an example of the generated sentence RGS after correction by the correction unit 14 in this step. Note that in the example in the lower part of Fig. 9, the bold type and underlining are mainly for the convenience of explanation in this specification, but as will be described later, they may also be used as decoration of the sentence when the generated sentence GS is presented to the user.

[0075] In this example, the correction unit 14 generates a corrected element RGSE7 by correcting the wording "if we talk about it" in element GS7, one of elements GSE6 to GSE8 contained in generated sentence GS1, to "if we consult with each other."

[0076] Here, the expression "if we discuss it" can be said to be an appropriate expression in normal conversation or documents, but it cannot be said to be an appropriate expression in the target business category of "contract" or "contract creation or editing work" in this example. In this step, the generated text GS is corrected using an industry-specific model that has been trained specifically for the target business category, so that detailed correction processing according to the business category can be performed with high accuracy, as described above.

[0077] (Step S124) In step S124, the control unit 10 outputs the generated sentence RGS after correction generated by the correction unit 14 as data via the input / output unit 40, for example, or presents it to the user. When presenting it to the user, the corrections made by the correction unit 14 may be presented with the corrections underlined, as shown in the lower part of FIG. 9, so that the user can recognize the corrections. Also, the changes from the elements included in the prompt PR1 may be shown in bold, as shown in the lower part of FIG. 9, so that the user can recognize the changes. Also, the control unit 10 may be configured to present both the generated sentence GS1 before correction and the generated sentence RGS1 after correction to the user.

[0078] In this manner, the information processing device 100 according to the present exemplary embodiment is configured to identify a target business category, generate a prompt corresponding to the identified business category, and obtain a generated sentence based on the prompt. Therefore, the information processing device 100 can provide a business support technology applicable to various businesses (for example, a technology for supporting a business of generating or editing a contract).

[0079] Furthermore, in the information processing device 100 according to this exemplary embodiment, an industry-specific model that is trained to be specialized for each business may be used in the process of extracting elements from the target sentence (step S112), the process of identifying the target business category (step S11), and the process of correcting the generated sentence GS (step S123). By using the industry-specific model, these processes can be performed with high accuracy.

[0080] (Example 2 of prompt generation and generated sentence output processing) 10 to 12, a prompt generation and generated sentence output processing example 2 executed by the information processing device 100 will be described. The processing according to this example includes processing similar to the above-mentioned "prompt generation and generated sentence output processing example 1", so the following description will focus on the differences between the processing according to this example and the above-mentioned "prompt generation and generated sentence output processing example 1".

[0081] Fig. 10 shows a user instruction IN2 which is an example of a user instruction acquired by the acquisition unit 13 according to this example in step S121 of Fig. 6. As shown in Fig. 10, the user instruction according to this example includes an instruction to check the target contract, "Check the previous contract."

[0082] The identification unit 11 according to this example identifies a task category by referring to the user instruction IN2 through a process similar to that of step S11 shown in Fig. 6. In this example, the identification unit 11 identifies that the target task category is "contract confirmation task" by referring to the words "contract" and "check" included in the user instruction IN2.

[0083] Next, the generation unit 12 according to this example acquires one or more elements associated with the target business category by the same process as step S122 shown in FIG. 6. Here, the generation unit 12 may acquire one or more elements associated with the target business category by further referring to the user instruction IN2. As an example, the generation unit 12 may be configured to identify the most recently processed contract by referring to the phrase "previous contract" included in the user instruction IN2, and acquire one or more elements associated with that contract. Note that these elements can be extracted in advance by the process described in "Element extraction process by the generation unit 12" above.

[0084] Next, the generation unit 12 according to this example generates a prompt PR by the same process as step S12 shown in Fig. 6. Fig. 11 shows a prompt PR2 that is an example of a prompt PR generated by the generation unit 12 according to this example with reference to a user instruction IN2.

[0085] As shown in Figure 11, prompt PR2 is · Tags that indicate imperative statements (#imperative statements) - Command statement content (PR2_main) · Tag indicating the input text (#input text) · Contents of input text (PR2_EM) · Tags that indicate points (points to note when checking) (#points) · Contents of points (points to note when checking) (PR2_CP1~PR2_CP3) Here, the command content (PR2_main) is the content generated by the generating unit 12 according to the present example in accordance with the task category identified by the identifying unit 11 according to the present example in step S11. In the example shown in Fig. 11, the command content (PR2_main) includes the command "You are a professional lawyer. Please check the following contract from a legal perspective and list any deficiencies in bullet points" corresponding to the target task category "contract review task."

[0086] On the other hand, the content of the input sentence (PR2_EM) includes one or more elements acquired by the generation unit 12 according to this example in step S122. In the example shown in Fig. 11, the elements associated with the "last contract" include the elements acquired by the generation unit 12 according to this example in step S122.

[0087] The contents of the points (points to note in the confirmation work) include several points to note in the relevant work category "contract confirmation work". For example, the contents of the points include the following general points to note PR2_CP1: · Is the content fair? Can you achieve your business objectives? - Does it contain any illegal content? It includes points to note such as:

[0088] In addition, the above points include the following points regarding the handling of confidential information: PR2_CP2 · Are the parties to the contract clear? · Is the purpose of disclosing confidential information accurate? ·Is the definition of confidential information clear? It includes points to note such as:

[0089] In addition, the above points include the following points regarding the handling of contract periods: PR2_CP3 Is the contract period of appropriate length? ·Are the automatic renewal conditions appropriate? etc. are included.

[0090] The generating unit 12 may be configured to generate the above-mentioned multiple points according to at least one of the target business category and the user instruction, or may be configured to select from multiple points of attention prepared in advance according to at least one of the target business category and the user instruction.

[0091] Alternatively, the generation unit 12 may generate a prompt template associated with the task category “contract confirmation task” as follows: · Tags that indicate imperative statements (#imperative statements) · Tag indicating the input text (#input text) · Tags that indicate points (points to note when checking) (#points) · Contents of points (points to note when checking) (PR2_CP1~PR2_CP3) The above-mentioned prompt PR may be generated by preparing a prompt template including the above and applying the contents of the command sentence (PR2_main) and the contents of the input sentence (PR2_EM) to the prompt template.

[0092] The prompt PR2 generated in this manner is provided to the generation device 50 by processing similar to that described in "Prompt Generation and Generated Sentence Output Processing Example 1," and the acquisition unit 13 in this example acquires the generated sentence GS2 generated by the language model LM based on the prompt PR2 in the generation device 50.

[0093] Fig. 12 is a diagram showing a generated sentence GS2 acquired by the acquiring unit 13 according to this example. As shown in Fig. 12, the generated sentence GS2 includes the confirmation results (GS2_EM1 to GS2_EM4) generated by the language model LM.

[0094] The confirmation result GS2_EM1 is a result generated in response to the general attention point PR2_CP1 in the prompt PR2. As an example, as shown in FIG. 12, The contract format is common. No grammatical errors are found. · No ambiguous terms are included. It includes confirmation results such as:

[0095] The confirmation result GS2_EM2 is a result generated in response to the precautions PR2_CP2 regarding the handling of confidential information in the prompt PR2. As an example, as shown in FIG. 12, The parties are clearly stated. · The purpose of disclosing the confidential information is clearly stated. · Confidential information is clearly defined. It includes confirmation results such as:

[0096] The confirmation result GS2_EM3 is a result generated in response to the consideration PR2_CP3 regarding the handling of the contract period in the prompt PR2. As an example, as shown in FIG. 12, The length of the contract is not specified, but that doesn't matter. -The terms of automatic renewal are clearly stated. It includes confirmation results such as:

[0097] The verification result GS2_EM4 shows the overall verification result obtained from the verification results from the multiple perspectives mentioned above. As an example, as shown in Figure 12, it includes the wording, "Based on the above, it has been confirmed that this contract does not contain any significant defects."

[0098] The generated sentence GS2 acquired by the acquisition unit 13 according to this example is presented to the user via the input / output unit 40, as one example. As another example, the correction unit 14 according to this example may be configured to correct the generated sentence GS2 and present the corrected generated sentence RGS2 to the user. In such a configuration, the correction process may use an industry-specific model specialized for the target business category, as in "Prompt Generation and Generated Sentence Output Processing Example 1."

[0099] In this processing example as well, the information processing device 100 is configured to identify a target business category, generate a prompt corresponding to the identified business category, and acquire a generated sentence based on the prompt. Therefore, the information processing device 100 can provide a business support technology applicable to various businesses (for example, a support technology for a contract confirmation business).

[0100] Furthermore, in the information processing device 100 according to this exemplary embodiment, an industry-specific model that is trained to be specialized for each business may be used in the process of extracting elements from the target sentence (step S112), the process of identifying the target business category (step S11), and the process of correcting the generated sentence GS (step S123). By using the industry-specific model, these processes can be performed with high accuracy.

[0101] In the above description, "Example 1 of the prompt generation and generated sentence output process" and "Example 2 of the prompt generation and generated sentence output process" are described separately, but this does not necessarily mean that these processes are performed separately. Generate a prompt including prompt PR1 and prompt PR2; The configuration may be such that generated sentences based on the prompts include a generated sentence GS1 based on prompt PR1 and a generated sentence GS2 based on prompt PR2.

[0102] [Software implementation example] Some or all of the functions of the information processing device 1,100 may be realized by hardware such as an integrated circuit (IC chip), or may be realized by software.

[0103] In the latter case, the information processing device 1,100 is realized, for example, by a computer that executes instructions of a program, which is software that realizes each function. An example of such a computer (hereinafter, referred to as computer C) is shown in FIG. 13. The computer C includes at least one processor C1 and at least one memory C2. The memory C2 stores a program P for operating the computer C as the information processing device 1,100. In the computer C, the processor C1 reads and executes the program P from the memory C2, thereby realizing each function of the information processing device 1,100.

[0104] The processor C1 may be, for example, a central processing unit (CPU), a graphic processing unit (GPU), a digital signal processor (DSP), a micro processing unit (MPU), a floating point number processing unit (FPU), a physics processing unit (PPU), a tensor processing unit (TPU), a quantum processor, a microcontroller, or a combination thereof. The memory C2 may be, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination thereof.

[0105] The computer C may further include a RAM (Random Access Memory) for expanding the program P during execution and for temporarily storing various data. The computer C may further include a communication interface for transmitting and receiving data to and from other devices. The computer C may further include an input / output interface for connecting input / output devices such as a keyboard, a mouse, a display, and a printer.

[0106] Furthermore, the program P can be recorded on a non-transitory tangible recording medium M that can be read by the computer C. Such a recording medium M can be, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit. The computer C can obtain the program P via such a recording medium M. Furthermore, the program P can be transmitted via a transmission medium. Such a transmission medium can be, for example, a communication network or broadcast waves. The computer C can also obtain the program P via such a transmission medium.

[0107] [Additional Note 1] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0108] (Appendix A1) A means for identifying a target business category; a generating means for generating a prompt according to the task category identified by the identifying means; an acquisition means for acquiring a generated sentence based on the prompt; An information processing device comprising:

[0109] (Appendix A2) The generating means includes: Generate the prompt including content according to an instruction from the user 2. The information processing device according to claim 1 .

[0110] (Appendix A3) The business category includes a contract creation or editing business, The prompt includes instructions to create or edit a contract. 13. The information processing device according to claim 12,

[0111] (Appendix A4) The above business categories include contract confirmation work, The prompt includes an instruction to review the agreement. 2. An information processing device according to claim 1,

[0112] (Appendix A5) The system further includes a correction means for correcting the generated sentence in accordance with the task category identified by the identification means. 6. An information processing device according to claim 1,

[0113] (Appendix A6) The generating means includes: executing an extraction process for extracting one or more elements corresponding to the task category identified by the identification means from one or more target sentences; Generate the prompt according to the element extracted in the extraction process. 2. The information processing device according to claim 1 ,

[0114] (Appendix A7) The generating means includes: generating the prompt including an instruction sentence corresponding to an instruction from a user and the element extracted by the extraction process; 2. The information processing device according to claim 1 ,

[0115] (Appendix A8) The generating means includes: The extraction process is performed using a language model trained for each task category. 2. The information processing device according to claim 1 ,

[0116] (Appendix B1) At least one processor Identifying a task category corresponding to the target sentence; generating a prompt according to the task category identified by the identifying means; Obtaining a generated sentence based on the prompt. An information processing method for performing the above.

[0117] (Appendix B2) In the generating step, Generate the prompt including content according to an instruction from the user 2. An information processing method as described in Appendix B1.

[0118] (Appendix B3) The business category includes a contract creation or editing business, The prompt includes instructions to create or edit a contract. 13. An information processing method according to claim 12, wherein the information processing method is a method for processing information according to claim 13.

[0119] (Appendix B4) The above business categories include contract confirmation work, The prompt includes an instruction to review the agreement. 10. The information processing method according to any one of appendices B1 to B3.

[0120] (Appendix B5) The method further includes a step of modifying the generated sentence according to the task category identified in the step of identifying. 10. An information processing method according to any one of appendices B1 to B4.

[0121] (Appendix B6) In the generating step, executing an extraction process for extracting one or more elements corresponding to the task category identified in the identifying step from one or more target sentences; Generate the prompt according to the element extracted in the extraction process. 2. An information processing method as described in Appendix B5.

[0122] (Appendix B7) In the generating step, generating the prompt including an instruction sentence corresponding to an instruction from a user and the element extracted by the extraction process; 2. An information processing method as described in Appendix B6.

[0123] (Appendix B8) In the generating step, The extraction process is performed using a language model trained for each task category. 2. An information processing method as described in Appendix B7.

[0124] (Appendix C1) On the computer, A process of identifying a business category corresponding to a target sentence; a generation process for generating a prompt according to the task category identified by the identification means; an acquisition process for acquiring a generated sentence based on the prompt; A program that executes the following.

[0125] (Appendix C2) The generation process includes: Generate the prompt including content according to an instruction from the user 2. The program described in Appendix C1.

[0126] (Appendix C3) The business category includes a contract creation or editing business, The prompt includes instructions to create or edit a contract. 1. A program according to appendix C1 or C2.

[0127] (Appendix C4) The above business categories include contract confirmation work, The prompt includes an instruction to review the agreement. 10. The program according to claim 9, wherein the program is a program for performing a method for producing a semiconductor device according to claim 9.

[0128] (Appendix C5) The method further includes a correction process for correcting the generated sentence according to the task category identified in the identification process. 10. The program according to claim 9, wherein the program is a program for performing a method for the production of a semiconductor device.

[0129] (Appendix C6) The generation process includes: executing an extraction process for extracting one or more elements corresponding to the task category identified in the identification process from one or more target sentences; Generate the prompt according to the element extracted in the extraction process. 2. The program described in Appendix C5.

[0130] (Appendix C7) The generation process includes: generating the prompt including an instruction sentence corresponding to an instruction from a user and the element extracted by the extraction process; 2. The program described in Appendix C6.

[0131] (Appendix C8) The generation process includes: The extraction process is performed using a language model trained for each task category. The program described in Appendix C7.

[0132] [Additional Note 2] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0133] (Appendix D1) at least one processor, the at least one processor comprising: A process of identifying a business category corresponding to a target sentence; a generation process for generating a prompt according to the task category identified by the identification means; an acquisition process for acquiring a generated sentence based on the prompt; An information processing device that executes the above.

[0134] The information processing device may further include a memory, and the memory may store a program for causing the at least one processor to execute each of the processes.

[0135] (Appendix D2) The generation process includes: Generate the prompt including content according to an instruction from the user 2. The information processing device according to claim 1 .

[0136] (Appendix D3) The business category includes a contract creation or editing business, The prompt includes instructions to create or edit a contract. The information processing device according to claim D1 or D2.

[0137] (Appendix D4) The above business categories include contract confirmation work, The prompt includes an instruction to review the agreement. 2. An information processing device according to claim 1, wherein the information processing device is a

[0138] (Appendix D5) The at least one processor: A correction process is further performed to correct the generated sentence according to the task category identified in the identification process. An information processing device according to any one of appendices D1 to D4.

[0139] (Appendix D6) The generation process includes: executing an extraction process for extracting one or more elements corresponding to the task category identified in the identification process from one or more target sentences; Generate the prompt according to the element extracted in the extraction process. 13. The information processing device according to claim 12,

[0140] (Appendix D7) The generation process includes: generating the prompt including an instruction sentence corresponding to an instruction from a user and the element extracted by the extraction process; 13. The information processing device according to claim 12,

[0141] (Appendix D8) The generation process includes: The extraction process is performed using a language model trained for each task category. 13. The information processing device according to claim 12, [Explanation of symbols]

[0142] 1,100 ···Information processing equipment 11...Specific section 12...Generation section 13...Acquisition part 14...Revised section

Claims

1. A means for identifying a target business category; a generating means for generating a prompt according to the task category identified by the identifying means; an acquisition means for acquiring a generated sentence based on the prompt; An information processing device comprising:

2. The generating means includes: Generate the prompt including content according to an instruction from the user The information processing device according to claim 1 .

3. The business category includes a contract creation or editing business, The prompt includes instructions to create or edit a contract. The information processing device according to claim 2 .

4. The above business categories include contract confirmation work, The prompt includes an instruction to review the agreement. The information processing device according to claim 2 .

5. The system further includes a correction means for correcting the generated sentence in accordance with the task category identified by the identification means. The information processing device according to claim 2 .

6. The generating means includes: executing an extraction process for extracting one or more elements corresponding to the task category identified by the identification means from one or more target sentences; Generate the prompt according to the element extracted in the extraction process. The information processing device according to claim 5 .

7. The generating means includes: generating the prompt including an instruction sentence corresponding to the instruction from the user and the element extracted by the extraction process; The information processing device according to claim 6 .

8. The generating means includes: The extraction process is performed using a language model trained for each task category. The information processing device according to claim 7.

9. At least one processor Identifying a task category corresponding to the target sentence; generating a prompt according to the task category identified in the identifying step; Obtaining a generated sentence based on the prompt. An information processing method for performing the above.

10. On the computer, A process of identifying a business category corresponding to a target sentence; a generation process for generating a prompt according to the task category identified in the identification process; an acquisition process for acquiring a generated sentence based on the prompt; A program that executes the following.

Images