Information processing system, information processing device, information processing method, and program

Abstract

A technique is provided that makes it possible to acquire desired output information. An information processing system obtains information related to an output format, searches for a plurality of pieces of reference information based on a search request, obtains one or more pieces of reference information to be used for generating output information based on a result of the search, generates input information for a machine learning model based on the information related to the output format and the one or more pieces of reference information, and generates the output information by inputting the input information into the machine learning model. and based on a result of the search, acquiring one or more other reference information to be used for generating second output information, generating second input information based on the information on the output format and the one or more other reference information, inputting the second input information into a machine learning model to generate second output information, and outputting difference information between the output information and the second output information. do.

Description

[Technical field]

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

[0002] Foundational models such as large-scale language models (LLMs) are known. Foundational models are machine learning models that can execute various tasks by learning large-scale data sets. Foundational models execute tasks according to input information called prompts, and output the resulting data.

[0003] In order to obtain good output results from the foundation model, a technique is known in which a prompt contains reference information obtained by searching a specific data source. This kind of technique is called Retrieval Augmented Generation (RAG). [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Kuttler, Mike Lewis, Wen-tau Yih, Tim Rocktaschel, Sebastian Riedel, Douwe Kiela, "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks", [online], [searched on December 20, 2020], Internet<URL: https: / / arxiv.org / pdf / 2005.11401v4.pdf> Summary of the Invention [Problem to be solved by the invention]

[0005] An object of the present disclosure is to provide a technology capable of acquiring desired output information. [Means for solving the problem]

[0006] An information processing system according to an embodiment of the present disclosure includes at least one memory and at least one processor, and the at least one processor acquires information related to an output format, searches for a plurality of reference information based on a search request, acquires one or more reference information used to generate output information based on a result of the search, generates input information for a machine learning model based on the information related to the output format and the one or more reference information, and generates the output information by inputting the input information to the machine learning model. and based on a result of the search, acquiring one or more other reference information to be used for generating second output information, generating second input information based on the information on the output format and the one or more other reference information, inputting the second input information into a machine learning model to generate second output information, and outputting difference information between the output information and the second output information. do. [Brief description of the drawings]

[0007] [Figure 1] 1 is a block diagram showing an example of an overall configuration of an information processing system. [Diagram 2] 2 is a block diagram showing an example of a functional configuration of a report device; FIG. [Diagram 3] FIG. 13 is a diagram illustrating an example of a template. [Figure 4] FIG. 13 is a diagram showing a first example of a format designation screen. [Diagram 5] FIG. 13 is a diagram showing a second example of a format designation screen. [Figure 6] FIG. 13 is a diagram showing an example of a condition setting screen. [Figure 7] FIG. 13 illustrates an example of a selection screen. [Figure 8] FIG. 4 is a diagram showing an example of a display screen. [Figure 9] 13 is a flowchart illustrating an example of a report generation process according to the first embodiment. [Figure 10] 10 is a flowchart illustrating an example of a report generation process according to the second embodiment. [Figure 11]FIG. 2 is a block diagram showing an example of a hardware configuration of a computer. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. In this specification and the drawings, components having substantially the same functional configurations are denoted by the same reference numerals, and redundant description will be omitted.

[0009] [First embodiment] A first embodiment of the present disclosure is an information processing system that generates various reports based on a base model. In this embodiment, the base model may be, for example, a large language model (LLM). The base model is not limited to a large language model, and may be a generative model that generates various data such as voice, image, and video. Note that this embodiment is not limited to the base model, and may be configured using a machine learning model. The machine learning model may include, for example, a base model, a generative model, or a neural network. The machine learning model may be multimodal.

[0010] The information processing system in this embodiment executes a predetermined generation task according to a search expansion generation framework. In this embodiment, the generation task may be, for example, a task of generating a report. For example, the report may be a report intended for information sharing within an organization such as a company, a local government, or a school, or a report intended for information provision outside the organization. For example, the report may be a research report that reports research results. The report may be a document that includes information desired by a user.

[0011] Search expansion generation is a technique used to obtain good generation results from a foundation model. In search expansion generation, reference information is searched for from a predetermined data source based on instruction information that instructs the execution of a task, and a prompt including the instruction information and the search results of the reference information is input to the foundation model. The foundation model, for example, executes a task while considering reference information that is not included in the training data, and therefore can generate data appropriate for the instruction information.

[0012] This embodiment provides a technology capable of acquiring desired output information in a search expansion generation framework. In this embodiment, the input information to the base model includes information on the output format and reference information searched based on the input information, thereby generating output information according to the output format. In one aspect, according to this embodiment, the output format can be changed by a simple operation, so that the desired output information can be easily acquired.

[0013] <Overall configuration of information processing system> The overall configuration of an information processing system in this embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the overall configuration of an information processing system.

[0014] 1, the information processing system 1000 includes a report device 10, a search device 20, a generation device 30, and a terminal device 40. The report device 10, the search device 20, the generation device 30, and the terminal device 40 may be connected to each other so as to be able to communicate data with each other via a communication network such as a LAN (Local Area Network) or the Internet.

[0015] The report device 10 is an example of an information processing device such as a personal computer, a workstation, or a server that instructs the generation device 30 to execute a predetermined task in response to a request from a terminal device 40. In this embodiment, the predetermined task may be a task of generating a report, which is an example of output information.

[0016] The report device 10 may obtain the search results of the reference information from the search device 20 based on the generation request received from the terminal device 40. The report device 10 may transmit a search request to the search device 20 based on the generation request.

[0017] A generation request is information or a signal that requests the generation of a report. A generation request may include information regarding the output format, input information entered by a user, and search conditions set by the user. Hereinafter, information regarding the output format will also be referred to as "output format information," and input information entered by a user will also be referred to as "user input information." Note that at least a portion of the "output format information" may be information entered by a user.

[0018] The search request is information or a signal requesting a search for reference information. The search request may include, by way of example, a query and a search condition. The query may include user-entered information included in the generation request. The search condition may include the search condition included in the generation request.

[0019] The report device 10 may generate input information for the base model 35 based on the search result of the reference information acquired from the search device 20. The report device 10 may transmit the search result of the reference information to the terminal device 40, and generate input information for the base model 35 based on the reference information selected by the terminal device 40.

[0020] The report device 10 may obtain output information generated by inputting input information to the base model 35 into the base model 35 from the generation device 30. The report device 10 may generate a report based on the output information and transmit it to the terminal device 40 as a generation result in response to the generation request.

[0021] The search device 20 is an example of an information processing device such as a personal computer, a workstation, or a server that searches for reference information in response to a request from the report device 10. The search device 20 may include a data source 25 that stores various reference information. The data source 25 may be, for example, a storage device or a database in which the reference information is stored. The data source 25 may be included in an external device or system. Here, "external" means that it is not included in the information processing system 1000.

[0022] The search device 20 may be implemented in multiple devices or systems with different data sources 25. The search device 20 may be implemented in one device or system with multiple data sources 25. The search device 20 may be an external device or system. The search device 20 may search for reference information from the external data sources 25.

[0023] The generating device 30 is an example of an information processing device such as a personal computer, a workstation, or a server that executes a predetermined task using a base model in response to a request from the report device 10. The generating device 30 may include a base model 35. The base model 35 is a machine learning model for executing the predetermined task. The base model 35 may be, for example, a large-scale language model.

[0024] The base model 35 may be realized by one base model. The base model 35 may be realized by cooperation of a plurality of base models. The base model 35 may be composed of a plurality of base models according to the tasks to be executed.

[0025] The generating device 30 may be realized in multiple devices or systems having different base models 35. The generating device 30 may be realized in one device or system having multiple base models 35. The generating device 30 may be an external device or system. The generating device 30 may execute a predetermined task based on the external base model 35.

[0026] The terminal device 40 is an example of an information processing terminal, such as a personal computer, a smartphone, or a tablet terminal, operated by a user of the information processing system 1000. The terminal device 40 may request the report device 10 to execute a predetermined task. The terminal device 40 may transmit a generation request including output format information and user input information to the report device 10. The terminal device 40 may receive a generation result from the report device 10 and present it to the user.

[0027] For example, the terminal device 40 may display the generation result on a display device of the terminal device 40. For example, the terminal device 40 may emit a voice synthesized from the generation result from a speaker of the terminal device 40.

[0028] Note that the overall configuration of the information processing system 1000 shown in FIG. 1 is just an example, and there may be various examples of the configuration of the information processing system depending on the application or purpose. The information processing system 1000 may be configured with one or more devices. The devices included in the information processing system 1000 may be a system configured with multiple devices. Each function included in the information processing system 1000 may be realized by any device that constitutes the system. Each component included in the information processing system 1000 may be included in any device that constitutes the system.

[0029] The data source 25 may be built in the report device 10 or the generation device 30. The data source 25 may be distributed and held in an external information processing system consisting of a plurality of devices. In this case, the information processing system 1000 may not include the search device 20.

[0030] The base model 35 may be built into the report device 10 or the search device 20. The base model 35 may be distributed and held in an external information processing system consisting of multiple devices. In this case, the information processing system 1000 may not include the generation device 30.

[0031] The information processing system 1000 may include a plurality of units of one or more of the report device 10, the search device 20, the generation device 30, and the terminal device 40. The report device 10, the search device 20, or the generation device 30 may be realized by a plurality of computers, or may be realized as a cloud computing service. Two or more of the report device 10, the search device 20, and the generation device 30 may be realized by a standalone computer. The division of devices such as the report device 10, the search device 20, the generation device 30, and the terminal device 40 shown in FIG. 1 is an example.

[0032] As an example, the information processing system 1000 may be configured with a server device and one or more terminal devices 40. The server device may include the functions of a report device 10, a search device 20, and a generation device 30. The server device may be realized as a system including a plurality of information processing devices. The server device may be realized as a cloud computing service.

[0033] As an example, the information processing system 1000 may be configured with a first server device, a second server device, and one or more terminal devices 40. The first server device may have the functions of a report device 10 and a generation device 30. The second server device may have the functions of a search device 20. The first server device and the second server device may each be realized as a system including a plurality of information processing devices. The first server device and the second server device may each be realized as a cloud computing service.

[0034] As an example, the information processing system 1000 may be configured with one or more information processing devices. The one or more information processing devices constituting the information processing system 1000 may have the functions of the report device 10, the search device 20, the generation device 30, and the terminal device 40. As another example, the information processing system may be configured with one or more information processing devices having the functions of the report device 10, the search device 20, and the generation device 30. The information processing system may also be configured with one or more information processing devices having the functions of the report device 10 and the generation device 30. The information processing system may also be configured with one or more information processing devices having the functions of the report device 10 and the generation device 30. The information processing system may also be configured with one or more information processing devices having at least some of the functions of the report device 10, the search device 20, the generation device 30, and the terminal device 40.

[0035] <Functional configuration of the report device> The functional configuration of the report device 10 will be described with reference to Fig. 2. Fig. 2 is a block diagram showing an example of the functional configuration of the report device.

[0036] 2, the report device 10 includes a request receiving unit 110, a search unit 120, a presentation unit 130, a selection receiving unit 140, a generation unit 150, a history storage unit 160, an output unit 170, and an accounting unit 180. The report device 10 functions as the request receiving unit 110, the search unit 120, the presentation unit 130, the selection receiving unit 140, the generation unit 150, the history storage unit 160, the output unit 170, and the accounting unit 180 by executing a program installed in advance.

[0037] The request receiving unit 110 receives a request to generate a report. The request receiving unit 110 may receive a generation request from the terminal device 40. The generation request may be transmitted from the terminal device 40 in response to an operation on a screen displayed on a display device of the terminal device 40. The request receiving unit 110 may receive a generation request input to the report device 10. The generation request may be input to the report device 10 in response to an operation on a screen displayed on a display device of the report device 10.

[0038] The generation request may include output format information and user input information. The output format information may include information specifying an output format. The output format information may include information regarding one or more output formats selected from a plurality of output formats. The output format information may be information capable of identifying the output format. The information capable of identifying the output format may, for example, be identification information that identifies the output format, or information indicating the location of a file describing the output format.

[0039] The output format information may include information regarding the structure of the report. The information regarding the structure of the report may be information indicating a structure of one or more items included in the report. The information indicating the structure may include, for example, information regarding at least one of the item names, the order of the items, or the hierarchy of the items. The structure of the report may include, for example, at least one of the title, subtitles, or table of contents of the report. The table of contents may include one or more headings.

[0040] The output format information may be information that specifies the format of the output information, which may include, for example, a natural language sentence, Markdown notation, HTML (Hyper Text Markup Language), XML (Extensible Markup Language), JSON (JavaScript Object Notation) format, CSV (Comma Separated Values) format, Mermaid format, etc.

[0041] The output format information may be information indicating a file format. For example, the information indicating the file format may be a file extension. For example, the file format may include text, image data, video data, audio data, a table format, a presentation format, or a portable document format (PDF).

[0042] The output format information may be selected from a plurality of output format information created in advance. The output format information may be newly created in response to an operation by a user. The output format information may be selected or generated in response to an operation on a screen displayed on the display device of the terminal device 40. As an example, the output format information may be written using a tool such as a text editor. The output format information created in advance may include a report created in the past.

[0043] The output format information may be, for example, a research report format, which may include items such as an overview, experiment details, results (discussion), and summary (conclusion).

[0044] The user-entered information may include search terms to be included in the query. The user-entered information may include, for example, the title of the report and headings to be included in the table of contents. The user-entered information may include, for example, one or more natural language sentences that describe the contents of the report as a whole or each item in the report.

[0045] The user input information may include text data, structured data, image data, or audio data. The image data may be, for example, a still image or a video. The text data may be text data generated by speech recognition of audio data or a video. The text data may be text generated by character recognition of image data. The audio data may be audio data generated by speech synthesis of the text data. The structured data may be, for example, a table or a graph.

[0046] The generation request may include search conditions for the reference information. The search conditions may include, for example, at least one of information indicating a data source to be searched, a type of the data source, a time range of a creation date, attribute information of the reference information to be searched, the number of items to be included in the search result, and a use condition.

[0047] The information indicating the data source may be information capable of identifying the data source. For example, the information capable of identifying the data source may be identification information for identifying the data source, or information indicating the location of the data source (e.g., a host name, an Internet Protocol (IP) address, a connection string, or a Uniform Resource Locator (URL)).

[0048] The information indicating the data source may be set for the entire report. The information indicating the data source may be set for each item included in the report. The information indicating the data source may include multiple data sources. The information indicating the data source may include a data source for the entire report and a data source for each item. When the information indicating the data source includes multiple data sources, it may include information indicating the priority of each data source.

[0049] The type of data source may include, for example, information indicating inside or outside of an organization. The type of data source inside an organization may include, for example, at least one of the entire organization, a department of an organization (e.g., a division of a company, etc.), or a business database (e.g., product information or sales information, etc.). The type of data source outside an organization may include, for example, at least one of a website (e.g., a news site or a dictionary site, etc.), a search engine, or a social networking service (SNS).

[0050] The attribute information of the reference information to be searched may include at least one of information indicating a file format, a creation date, and a usage condition. The information indicating a file format may be a file extension, for example. The usage condition may include billing information regarding the use of the reference information. The billing information may be information indicating whether or not to search for paid reference information.

[0051] The generation request may include user information regarding the user. The user information may include, for example, at least one of the user's authority, attributes, affiliation, position, or usage plan. The user's authority may include, for example, information indicating whether the user is a member of the management level of an organization. The usage plan may include, for example, information indicating whether paid reference information is available.

[0052] The search unit 120 searches for reference information based on the generation request accepted by the request acceptance unit 110. As an example, the search unit 120 may transmit a search request including user input information included in the generation request to the search device 20. The search unit 120 may receive search results transmitted by the search device 20 in response to the search request. The search results may include multiple pieces of reference information that satisfy the search conditions indicated in the search request.

[0053] The search request may include at least one of a query and a search condition. The query may be generated based on user input information included in the generation request. As an example, the query may include a keyword included in the user input information. The search condition may include a search condition included in the generation request. The search condition may include a predetermined search condition. The search unit 120 may set the search condition based on the user information included in the generation request.

[0054] The search unit 120 may search for reference information from multiple data sources indicated in the search conditions. The search unit 120 may control the data sources to be searched based on the user information included in the generation request. As an example, when the user's authority or usage plan indicates that paid reference information cannot be used, the search unit 120 may exclude data sources in which paid reference information is stored from the search targets.

[0055] The search unit 120 may include in the search results the number of reference information items indicated in the search conditions. The search unit 120 may include in the search results a predetermined number of reference information items. The search unit 120 may set the number of reference information items to be included in the search results based on the user's authority or usage plan. The search unit 120 may set the number of reference information items to be included in the search results based on the amount of data that can be input to the base model 35.

[0056] The presentation unit 130 presents to the user information on the search results of the reference information searched by the search unit 120. The information on the search results may include, for example, at least one of the data source of the reference information, link information, at least a portion of the information, a summary, a thumbnail image, an owner, a copyright holder, a creation date, or usage conditions. The usage conditions may include information indicating whether or not the use of the reference information is for a fee. The information indicating whether or not the use is for a fee may include information indicating whether the use of the reference information is for a fee or free of charge, a usage fee, or consumed points. The usage fee and consumed points may be numerical values, and may be 0 if free of charge.

[0057] The presentation unit 130 may transmit the search result of the reference information to the terminal device 40 and display the search result of the reference information on the display device of the terminal device 40. The presentation unit 130 may display the search result of the reference information on the display device of the report device 10. Note that presenting information to the user may include the processor executing at least a part of the processing required for displaying information on the display device. The display device may be provided in the same device as the processor, or in a device different from the processor. The display device may be a plurality of display devices.

[0058] The presentation unit 130 presents the search results of the reference information in a manner that allows the user to select it. The presentation unit 130 may control the selectable reference information based on the user information. As an example, when the user information indicates that the paid reference information is not available, the presentation unit 130 may present the paid reference information to the user in a manner that makes it unselectable.

[0059] The presentation unit 130 may present information about the search results (hereinafter, also referred to as auxiliary information of the reference information) together with the search results of the reference information. The auxiliary information of the reference information may include, for example, at least one of a data source, link information, partial information, summary, thumbnail image, owner, copyright holder, creation date, or usage conditions. The auxiliary information of the reference information may present any combination of two or more of these. The presentation unit 130 may generate at least a part of the auxiliary information of the reference information based on the base model 35 or another base model.

[0060] The data source is information indicating the data source in which the reference information was stored. The link information is information indicating the location of the reference information (e.g., a URL, etc.). The portion of information may include, for example, at least one of the title or heading of the reference information, a predetermined number of characters or a predetermined number of lines from the beginning, or keywords extracted from the reference information. For example, the summary or thumbnail image may be obtained from the data source 25, or may be generated based on the foundation model 35 or another foundation model.

[0061] The usage conditions may include billing information. The billing information may include information indicating whether the reference information can be used for a fee or free of charge. The usage conditions may include copyright information. The copyright information may include terms of use. The usage conditions may be set by the owner or copyright holder of the reference information. When the usage conditions of the reference information indicate that the reference information cannot be used, the presentation unit 130 may present the user with a message encouraging the user to enter into a contract with the owner or copyright holder of the reference information.

[0062] The selection receiving unit 140 receives a selection request for reference information. The selection receiving unit 140 may receive a selection request from the terminal device 40. The selection request may be transmitted from the terminal device 40 in response to an operation on a screen displayed on a display device of the terminal device 40. The selection receiving unit 140 may receive a selection request input to the report device 10. The selection request may be input to the report device 10 in response to an operation on a screen displayed on a display device of the report device 10.

[0063] The selection request is information or a signal requesting the selection of one or more pieces of reference information to be used in generating the output information. The selection request may be information indicating one or more pieces of reference information selected by the user from the search results of the reference information presented by the presentation unit 130.

[0064] The selection receiving unit 140 acquires the reference information indicated in the selection request. The selection receiving unit 140 may acquire information capable of identifying the reference information. The information capable of identifying the reference information may be, for example, identification information for identifying the reference information, or information indicating the location of a file in which the reference information is described.

[0065] The selection request may include one or more pieces of reference information specified by a user. The reference information specified by a user may be reference information that is not included in the search results of the reference information. As an example, the reference information specified by a user may be information indicating the location of the reference information (e.g., a URL, etc.).

[0066] The generation unit 150 acquires output information from the base model 35 based on the generation request accepted by the request accepting unit 110 and one or more pieces of reference information acquired by the selection accepting unit 140. As an example, the generation unit 150 may generate input information for the base model 35 based on the generation request and the reference information, and transmit the information to the generation device 30. Hereinafter, the input information for the base model 35 is also referred to as "model input information."

[0067] The generating unit 150 may generate model input information by acquiring output format information and user input information from the generation request and embedding the output format information, user input information, and reference information in a predetermined template. The generating unit 150 may generate model input information by processing the template, output format information, user input information, and reference information based on a predetermined rule. The generating unit 150 may generate model input information by inputting the template, output format information, user input information, and reference information to another base model. The generating unit 150 may receive output information from the generating device 30 that is generated by the generating device 30 inputting the model input information to the base model 35.

[0068] The generating unit 150 may generate the model input information based on one or more pieces of reference information searched by the searching unit 120, one or more pieces of reference information selected by the user, or one or more pieces of reference information designated by the user. The generating unit 150 may generate the model input information based on any combination of one or more pieces of reference information searched by the searching unit 120, one or more pieces of reference information selected by the user, and one or more pieces of reference information designated by the user. The generating unit 150 may generate the model input information using at least a part of the information included in each piece of reference information. The generating unit 150 may also generate the model input information using information obtained by processing each piece of reference information. The generating unit 150 may also generate the model input information using at least a part of the information included in the output format information, or may generate the model input information using information obtained by processing the output format information. The generating unit 150 may also generate the model input information using at least a part of the information included in the user input information, or may generate the model input information using information obtained by processing the user input information.

[0069] The generation unit 150 may generate output information by inputting model input information to one base model 35 selected from the multiple base models 35. The base model 35 used to generate the output information may be specified by a generation request, for example. The base model specified by the generation request may be selected by a user. The base model 35 used to generate the output information may be selected based on the user's authority or usage plan, for example.

[0070] A template for generating model input information may include one or more placeholders for embedding output format information, user input information, or reference information. The template may include instruction information for instructing the generation of a report. The template may include information specifying an output format. The template may include constraint information regarding the output information. The instruction information and constraint information may be predetermined fixed sentences.

[0071] The model input information may include text data, image data, or audio data. The text data may be, for example, a natural language sentence called a prompt. The image data may be, for example, a still image or a video. The text data may be text data generated by speech recognition of audio data or a video. The text data may be text generated by character recognition of image data. The image data may include, for example, an image captured of the user. The audio data may include, for example, the voice spoken by the user. The audio data may be audio data generated by speech synthesis of the text data.

[0072] The history storage unit 160 stores history information of the output information generated by the generation unit 150. The output information may be accumulated in chronological order in the history storage unit 160. The history storage unit 160 may store information used in generating the output information. The information used in the generation may include, for example, at least one of a data source to be searched, a base model used in the generation, user input information, output format information, reference information used in the generation, or model input information.

[0073] The output information and the information used for generation may be stored in association with each other in the history storage unit 160. As an example of a method for associating the output information and the information used for generation, the output information and the information used for generation may be stored as one set, information that allows one piece of information to be acquired from the other piece of information may be stored, or one piece of information and identification information for the other piece of information may be stored as one set.

[0074] The output unit 170 outputs a generation result in response to the generation request accepted by the request accepting unit 110, based on the output information acquired by the generation unit 150. The output unit 170 may transmit the generation result including the output information to the terminal device 40. The output unit 170 may display the generation result including the output information on the display device of the report device 10.

[0075] The output unit 170 may generate a report based on the output information and output the generation result including the report. The output unit 170 may generate the output information itself as a report. The output unit 170 may generate a report including a part of the output information. The output unit 170 may generate a report by processing the output information.

[0076] The output unit 170 may include, in the report, auxiliary information of the reference information used to generate the output information. The auxiliary information of the reference information may be the same as the auxiliary information displayed by the presentation unit 130 together with the search results of the reference information.

[0077] The output unit 170 may output the report in a predetermined file format so as to be savable. The output unit 170 may output the report in an editable form. The output unit 170 may output the edited report in a predetermined file format so as to be savable. The file format may be selectable by a user. As an example, the output unit 170 may output output information written in Markdown notation so as to be editable. The output unit 170 may output the Markdown notation so as to be extensible.

[0078] The output unit 170 may set a usage range of the report. The usage range may include at least one of the number of copies or the range of disclosure, for example. The output unit 170 may determine the usage range of the report based on the usage conditions of the reference information used to generate the report.

[0079] The output unit 170 may display a plurality of reports in a comparable manner. The output unit 170 may display a plurality of reports arranged vertically or horizontally. The plurality of reports may each include output information acquired by the generation unit 150 and output information stored in the history storage unit 160. The plurality of reports may include a plurality of pieces of output information acquired by the generation unit 150 at different times.

[0080] The multiple reports may be multiple reports generated based on different information. The multiple reports may be multiple reports using different base models to generate output information, for example. The multiple reports may be multiple reports using different data sources as search targets, for example. The multiple reports may be multiple reports using different user input information, for example. The multiple reports may be multiple reports using different base models, data sources, or two or more of different user input information, for example.

[0081] The output unit 170 may output difference information of the multiple reports displayed for comparison. The output unit 170 may highlight the differences between the multiple reports. For example, the highlighting may be performed by changing at least one of the character type, character color, background color, and size.

[0082] The charging unit 180 charges the user for the cost of generating the report. The cost may be in the form of currency or points. As an example, charging in currency may be settled each time, or the charged amount may be settled in a lump sum at a predetermined time interval (e.g., one month, etc.). As an example, charging in points may be settled by consuming points purchased in advance by the user, or points granted in advance to the user.

[0083] When the search unit 120 searches for a data source storing reference information for which a fee is charged, or when the selection receiving unit 140 receives a selection of reference information for which a fee is charged, the charging unit 180 may charge the user for the cost of generating the report. The charging unit 180 may determine the cost to be charged to the user based on at least one of the amount and number of reference information used to generate the report. The charging unit 180 may control the search unit 120, the selection receiving unit 140, or the generation unit 150 so as not to use more than the number of reference information specified in the user's usage plan.

[0084] The charging unit 180 may charge the user for the cost of using the report. The charging unit 180 may determine the cost required for generating or using the report based on the conditions of use of each piece of reference information used to generate the report. The use of the report may include, for example, making the report available for the user to check, downloading the report, or copying the report. The charging unit 180 may charge the user for the cost of using the report when the user displays the report or saves the report. The charging unit 180 may determine the cost to be charged to the user based on at least one of the number of times the report is displayed or the number of times the report is saved.

[0085] The billing unit 180 may present the user with additional information regarding the generation or use of the report, including at least one of the time, cost, or computing resources required to generate or use the report.

[0086] The billing unit 180 may present additional information regarding the generation of the report to the user before outputting the report. The billing unit 180 may present a cost required for generating the report to the user before outputting the report. For example, the billing unit 180 may present the additional information including the cost to the user before the search unit 120 searches a data source that stores reference information for which a fee is required, or before the generation unit 150 generates model input information using reference information for which a fee is required.

[0087] When the accounting unit 180 acquires a generation instruction from the user after presenting the user with additional information regarding report generation, the accounting unit 180 may notify the generation unit 150 that the generation instruction has been acquired. When the generation unit 150 is notified by the accounting unit 180 that the generation instruction has been acquired, the generation unit 150 may generate model input information and transmit it to the generation device 30. By presenting the cost to the user before generating the report, the user can check the cost required for generation before generating the report and then command the execution of the generation process. If the cost is higher than the user expected, the user can cancel the generation process.

[0088] After outputting the report, the billing unit 180 may present the user with additional information regarding the use of the report. For example, the billing unit 180 may present the user with additional information, including the cost, before the user views the report or before the user saves the report.

[0089] 2 is an example, and various configuration examples are possible depending on the application and purpose. The report device 10 may not include the presentation unit 130 and the selection reception unit 140. In this case, the generation unit 150 may generate model input information based on the search result of the reference information searched by the search unit 120. As an example, the generation unit 150 may generate model input information by inputting all reference information or a predetermined number of reference information included in the search result into a template.

[0090] There may be more than one user operating the terminal device 40. For example, the user who inputs the output format information and the user input information, the user who selects the reference information, and the user who refers to the generated result may be the same user or different users. The terminal device that performs the operation of inputting the output format information and the user input information, the terminal device that performs the operation of selecting the reference information, and the terminal device that displays the generated result may be the same terminal device or different terminal devices.

[0091] <Prompt template> A template for generating a prompt, which is an example of model input information, will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of the template.

[0092] 3, the template 300 includes instruction information 301, source information 302, search result 303, instruction information 304, and output format 305. The instruction information 301 and 304 include information for instructing the generation of a report and constraint information regarding the generation of the report. The instruction information 301 and 304 may be predetermined fixed sentences.

[0093] The source information 302 includes a placeholder {info}. The placeholder {info} contains information about the data source that stored the reference information used to generate the output information. The search result 303 includes a placeholder {context}. The placeholder {context} contains information about the data source that stored the reference information used to generate the output information.

[0094] The output format 305 includes output format information. The output format 305 includes placeholders {title}, {prompt}. The placeholder {title} is filled with user-entered information entered in the title of the report. The placeholder {prompt} is filled with user-entered information entered in the heading of the report.

[0095] Although the template 300 includes one placeholder {context} for embedding reference information, the template may include multiple placeholders for embedding reference information. Each of the multiple placeholders may be embedded with multiple pieces of reference information. Note that at least a part of the information included in the reference information may be embedded in each placeholder, or information obtained by processing the reference information may be embedded.

[0096] The generation unit 150 can generate model input information by embedding necessary information in a placeholder of a template. The generation unit 150 may generate model input information without using a template. However, the generation unit 150 can easily generate model input information by using a template. The template may be optimized in advance to obtain good output information. By generating a prompt using an optimized template, good output information can be stably obtained from the foundation model 35.

[0097] <User Interface> The user interface provided by the report device 10 will be described with reference to Fig. 4 to Fig. 7. As an example, the user interface of the report device 10 is displayed on a display device of the report device 10 or the terminal device 40. In this embodiment, an example in which each screen is displayed on the display device of the terminal device 40 will be described.

[0098] The user interface of the report device 10 may include, for example, a format designation screen, a condition setting screen, a selection screen, and a display screen. The format designation screen is a screen for designating the output format of the report. The condition setting screen is a screen for setting search conditions for reference information. The selection screen is a screen for selecting reference information to be used in generating the report. The display screen is a screen for displaying the report.

[0099] ≪Format specification screen≫ 4 is a diagram showing a first example of a format designation screen. The first example of the format designation screen is an example of a format designation screen for designating the structure of a report.

[0100] As shown in FIG. 4, the format designation screen 600 has a style selection section 601 , a new creation button 602 , a title input section 603 , a hierarchy selection section 604 , a heading input section 605 , a delete link 606 , an add link 607 , a set button 608 and a start button 609 .

[0101] The style selection section 601 displays a list of report styles that have been created. A report style is an example of output format information. A report style is information that defines the structure of a report. The report structure includes a title and a table of contents. When a user selects a report style in the style selection section 601, the title and table of contents defined in the report style are displayed on the format designation screen 600.

[0102] The new button 602 is a button for creating a new report style. When the user presses the new button 602, the title and table of contents displayed on the format designation screen 600 are initialized, and input becomes possible.

[0103] The title input section 603 is an input field for inputting the title of the report. The title input section 603 may accept input of the report title by the user. The title input section 603 may initially display a title set in the report style selected in the style selection section 601. The title input section 603 may accept editing of the initially displayed title. The title initially displayed in the title input section 603 or a title edited by the user is embedded in the placeholder {title} of the template 300 shown in FIG. 3.

[0104] The layer selection section 604 is a selection field for selecting the layer of each item included in the table of contents. H2, H3, and H4 displayed in the layer selection section 604 are symbols indicating the heading layers. The layer selection section 604 may be added or deleted in response to a user operation.

[0105] The heading input section 605 is an input field for inputting the heading of each item included in the table of contents. The heading input section 605 may accept input of the heading of the item by the user. The heading input section 605 may initially display a heading set in the report style selected in the style selection section 601. The heading input section 605 may accept editing of the initially displayed heading. The heading initially displayed in the heading input section 605 or a heading edited by the user is embedded in the placeholder {prompt} of the template 300 shown in FIG. 3.

[0106] The delete link 606 is an operation section for deleting the heading input section 605. When the user operates the delete link 606, the layer selection section 604 and the heading input section 605 corresponding to the delete link 606 are deleted from the format designation screen 600.

[0107] The add link 607 is an operation section for adding a heading input section 605. When the user operates the add link 607, a new layer selection section 604 and heading input section 605 are added to the format designation screen 600.

[0108] The setting button 608 is a button for setting search conditions. When the user presses the setting button 608, a condition setting screen is launched.

[0109] The start button 609 is a button for starting the generation of a report. When the user presses the start button 609, a generation request is sent to the report device 10 based on the information entered on the format designation screen 600.

[0110] 5 is a diagram showing a second example of a format designation screen. The second example of the format designation screen is an example of a format designation screen for designating a file format of a report.

[0111] As shown in FIG. 5, a format designation screen 610 has a style selection section 601 , a new creation button 602 , a file format selection section 611 , an output item selection section 612 , a setting button 608 , and a start button 609 .

[0112] The file format selection section 611 displays options of file formats for the report in a selectable manner. The options of the file formats may be predetermined. The options of the file formats may include template files that have been created in advance for each file format. The structure of the report may be described in the template file. As an example, the title and headings may be described in the template file.

[0113] The output item selection section 612 displays options for the output items of the report in a selectable manner. The options for the output items may be determined in advance or may be read from a template file. The output items selected by the user in the output item selection section 612 may be included in the output format information as a table of contents of the report, for example.

[0114] <Condition setting screen> Fig. 6 is a diagram showing an example of a condition setting screen. As shown in Fig. 6, a condition setting screen 620 has a category selection section 621, a data source input section 622, a fee setting section 623, a type input section 624, a period input section 625, a delete link 626, an add link 627, and a confirm button 628.

[0115] The category selection section 621 displays options for the category of the data source in a selectable manner. The category options may include, for example, inside the organization and outside the organization.

[0116] The data source input section 622 is an input field for inputting a data source. The data source inputted in the data source input section 622 is used for searching for reference information. When a user inputs the name of a data source in the data source input section 622, a predetermined name of the data source may be supplemented. Note that the data source input section 622 is an input field for setting a data source for the entire report, but a data source for each item of the report may also be set.

[0117] The fee setting section 623 displays whether or not to search for paid reference information for each data source so that the user can select it. When searching for paid reference information is selected in the fee setting section 623, reference information is searched for from the data source regardless of whether it is paid or free. On the other hand, when not searching for paid reference information is selected in the fee setting section 623, only free reference information is searched from the data source. Note that the fee setting section 623 may be displayed only for data sources external to the organization.

[0118] The type input section 624 is an input field for inputting the file format of the reference information. The type input section 624 may also accept input of a file extension. When a file format is input into the type input section 624, only reference information in that file format is subject to search.

[0119] The period input section 625 is an input field for inputting a time range of reference information. The period input section 625 may accept input of both or either of the start and end of the time range to be searched. When a time range is input in the period input section 625, only reference information whose creation date falls within the time range will be searched.

[0120] The delete link 626 is an operation section for deleting the data source input section 622. When the user operates the delete link 626, the category selection section 621 and the data source input section 622 corresponding to the delete link 626 are deleted from the condition setting screen 620.

[0121] The add link 627 is an operation section for adding a data source input section 622. When the user operates the add link 627, a new category selection section 621 and a data source input section 622 are added to the condition setting screen 620.

[0122] The Confirm button 628 is a button for confirming the setting of the search conditions. When the user presses the Confirm button 628, the search conditions are set based on the information entered in the condition setting screen 620. The set search conditions are included in a generation request sent to the report device 10.

[0123] On the condition setting screen 620, information that can be set in the category selection section 621, data source input section 622, type input section 624, and period input section 625 may be controlled based on the user's authority or usage plan.

[0124] Here, an example has been described in which the condition setting screen 620 is activated by an operation on the format designation screens 600 and 610, but the condition setting screen 620 may be activated at any timing. For example, the condition setting screen 620 may be activated before the format designation screens 600 and 610 are activated. In other words, the search conditions may be set before the output format information and the user input information are input.

[0125] <Selection screen> 7 is a diagram showing an example of the selection screen 630. When the terminal device 40 receives the search results for the reference information from the report device 10, the selection screen 630 is displayed on the display device of the terminal device 40.

[0126] As shown in FIG. 7, the selection screen 630 has a list display section 631 , a create button 632 , and a search again button 633 .

[0127] The list display unit 631 displays search results for reference information. The list display unit 631 displays reference information included in the search results in a selectable manner. The list display unit 631 displays auxiliary information of the reference information together with information indicating the reference information. The list display unit 631 may display the title, summary, data source, creation date, owner, copyright holder, and terms of use (paid or free) of each piece of reference information. The title is an example of information indicating the reference information. The summary, data source, creation date, owner, copyright holder, and terms of use are examples of auxiliary information. At least a portion of the auxiliary information of the reference information may be generated based on the foundation model 35 or another foundation model.

[0128] The list display section 631 may display link information of the reference information. The user may click on the link information to check the entire text of the reference information. The list display section 631 may display thumbnails of the reference information.

[0129] The list display unit 631 may display information indicating whether each piece of reference information is selected. The information indicating whether each piece of reference information is selected may be, for example, at least one of a check box, a check mark, a background color, or a text color. The background color or text color may be displayed in a different color depending on whether each piece of reference information is selected.

[0130] The list display unit 631 may accept input of new reference information. The new reference information may be specified by a user of the terminal device 40 displaying the selection screen 630. The selection screen 630 may accept input of new reference information on a screen component different from the list display unit 631. The new reference information may be input on a screen different from the selection screen 630, and the selection screen 630 may accept an operation to launch that screen. The list display unit 631 may accept input of auxiliary information together with the new reference information. At least a part of the auxiliary information of the new reference information may be generated based on the foundation model 35 or another foundation model.

[0131] The list display unit 631 may have an add button for adding new reference information. The user may add new reference information by pressing the add button. The user may add reference information by simply specifying link information of the reference information. The list display unit 631 may generate and display a title, summary, etc. of the new reference information based on the information of the link destination.

[0132] A user may add new reference information by moving a file to a predetermined location by dragging and dropping the file. In this case, the information contained in the file becomes the new reference information. The predetermined location to which the file is moved may be the list display section 631, or may be newly set in the selection screen 630. As an example, the reference information may be added by moving the file into the frame of the list display section 631. The object to be dragged and dropped is not limited to a file, and may be an object that is generally subject to a drag-and-drop operation, such as text, an image, or link information. Even if an object other than a file is dragged and dropped, the information contained in the object or the information linked to the object becomes the new reference information. A user may add new reference information by storing a file to be used as reference information in a predetermined directory. The predetermined directory may be a directory designated by the user, or may be a directory assigned to each user by the information processing system.

[0133] New reference information added by the user may be displayed in a manner that allows it to be distinguished from other reference information in the list display area 631. As an example, for new reference information added by the user, information indicating that fact may be displayed in the "Data Source" field. For reference information added by dragging and dropping a file, the file name may be displayed in the "Title" field.

[0134] The list display unit 631 may display information for specifying new reference information. The information for specifying reference information may be, for example, information about drag and drop, or information about a directory for storing new reference information. The information about drag and drop may include, for example, information about a destination of drag and drop (which may be the destination frame itself), or an indication (which may be an explanatory text) that addition is possible by drag and drop. The information for specifying reference information may be information for separately specifying reference information included in multiple pieces of reference information by drag and drop. In other words, the information for specifying reference information does not have to be information for specifying only new reference information.

[0135] The generate button 632 is a button for instructing the generation of output information. When the user presses the generate button 632, a selection request indicating the reference information selected in the list display section 631 is transmitted to the report device 10.

[0136] The search again button 633 is a button for searching for reference information again. When the user presses the search again button 633, the selection screen 630 closes and returns to the format designation screen 600. At this time, the search results of the reference information displayed in the list display area 631 may be discarded or may be retained. If the search results of the reference information are retained, the two search results may be displayed in a comparable manner on the selection screen 630 displayed after the re-search.

[0137] ≪Display screen≫ 8 is a diagram showing an example of a display screen 640. When the terminal device 40 receives the generated result from the report device 10, a display screen 640 is displayed on the display device of the terminal device 40.

[0138] As shown in FIG. 8, the display screen 640 has a report display section 641 , a source display section 642 , a regenerate button 643 , a reselect button 644 , and a save button 645 .

[0139] The report display unit 641 displays the report received from the report device 10. The report display unit 641 may display the report in an editable manner. For example, the report display unit 641 may have a function of a text editor.

[0140] The source display section 642 displays source information of the reference information used to generate the report. The source information may include, for example, a link to the reference information, a data source that stored the reference information, and a condition for using the reference information (paid or free). When a user operates the link to the reference information, the user may be able to view the reference information.

[0141] The regenerate button 643 is a button for regenerating a report. When the user presses the regenerate button 643, the display screen 640 closes and returns to the format designation screen 600. When returning to the format designation screen 600, the user can change the output format or user input information and start generating a new report.

[0142] The reselect button 644 is a button for reselecting reference information. When the user presses the reselect button 644, the display screen 640 closes and returns to the selection screen 630. When the user returns to the selection screen 630, he or she can change the selected reference information and start generating a new report.

[0143] The save button 645 is a button for saving a report. When a user presses the save button 645, the report displayed in the report display section 641 is stored in the storage device of the terminal device 40. When a user edits a report in the report display section 641, the edited report is saved. Whether the save button 645 can be pressed may be controlled according to the user's authority or usage plan.

[0144] <Other screens> The user interface of the report device 10 may include a screen for selecting a foundation model to be used for generating a report from a plurality of foundation models. The user interface of the report device 10 may include a screen for selecting a data source to be used for searching for reference information from a plurality of data sources. The user interface of the report device 10 may include a screen for setting a selectable data source or foundation model.

[0145] The user interface of the report device 10 may include a screen for confirming the fee charged to the user.The user interface of the report device 10 may include a screen for settling the fee charged to the user.

[0146] <Report generation process flow> The report generation process executed by the information processing system 1000 will be described with reference to Fig. 9. Fig. 9 is a flowchart showing an example of the report generation process according to the first embodiment.

[0147] In step S1, the terminal device 40 transmits a generation request to the report device 10 in response to a user's operation on the format designation screen 600. The generation request includes the output format information and user input information input on the format designation screen 600. The generation request may also include search conditions input on the condition setting screen 620.

[0148] The request receiving unit 110 of the report device 10 receives a generation request from the terminal device 40. The request receiving unit 110 accepts the received generation request. The request receiving unit 110 sends user input information and search conditions to the search unit 120. The request receiving unit 110 also sends output format information and user input information to the generation unit 150.

[0149] In step S2, the search unit 120 of the report device 10 receives the user input information and the search conditions from the request receiving unit 110. The search unit 120 transmits a search request to the search device 20 based on the user input information and the search conditions. The search request includes a query including the user input information and the search conditions.

[0150] The search device 20 receives a search request from the report device 10. The search device 20 searches for reference information from the data source 25 based on the search request. The search device 20 transmits the search result for the reference information to the report device 10. Note that the search device 20 may search for reference information by any method based on the search request.

[0151] The search unit 120 of the report device 10 receives the search results of the reference information from the search device 20. The search unit 120 sends the search results of the reference information to the presentation unit 130.

[0152] In step S3, the presentation unit 130 of the report device 10 receives the search results for the reference information from the search unit 120. The presentation unit 130 transmits the search results for the reference information to the terminal device 40.

[0153] The terminal device 40 receives the search results of the reference information from the report device 10. The terminal device 40 displays a selection screen 630 on the display device. The selection screen 630 displays the search results of the reference information.

[0154] In step S4, the terminal device 40 determines whether to search again for reference information. Specifically, when the search again button 633 on the selection screen 630 is pressed, the terminal device 40 determines to search again for reference information. On the other hand, when the generate button 632 on the selection screen 630 is pressed, the terminal device 40 determines not to search again for reference information.

[0155] If the reference information is to be searched again (YES), the terminal device 40 requests the report device 10 to search again. In response to the request from the terminal device 40, the report device 10 returns the process to step S2. On the other hand, if the reference information is not to be searched again (NO), the terminal device 40 transmits a selection request to the report device 10. The selection request includes information indicating the reference information selected on the selection screen 630.

[0156] In step S5, the selection receiving unit 140 of the report device 10 receives a selection request from the terminal device 40. The selection receiving unit 140 accepts the received selection request. The selection receiving unit 140 acquires reference information indicated in the selection request. The selection receiving unit 140 sends the acquired reference information to the generation unit 150. The selection receiving unit 140 may also send new reference information added by the user to the generation unit 150.

[0157] The processes from step S3 to step S5 may be omitted. In this case, in step S2, the search unit 120 may send the search result of the reference information received from the search device 20 to the generation unit 150. Also, the selection receiving unit 140 may send new reference information added by the user to the generation unit 150.

[0158] In step S6, the generation unit 150 of the report device 10 receives the reference information from the selection receiving unit 140. The generation unit 150 also receives the output format information and the user input information from the request receiving unit 110.

[0159] The generating unit 150 generates the model input information based on the output format information, the user input information, and the reference information. The generating unit 150 may generate the model input information by embedding the output format information, the user input information, and the reference information in a predetermined template. The generating unit 150 transmits the model input information to the generating device 30.

[0160] In step S7, the generating device 30 receives model input information from the reporting device 10. The generating device 30 inputs the received model input information to the base model 35. The base model 35 executes a predetermined task based on the model input information, and outputs output information generated as a result. The generating device 30 transmits the output information output from the base model 35 to the reporting device 10.

[0161] The generation unit 150 of the report device 10 receives output information from the generation device 30. The generation unit 150 stores the received output information in the history storage unit 160 in association with information used to generate the output information. The generation unit 150 also sends the output information to the output unit 170.

[0162] In step S8, the output unit 170 of the report device 10 receives the output information from the generation unit 150. The output unit 170 generates a report based on the output information. The output unit 170 transmits the generation result including the generated report to the terminal device 40. The report may include auxiliary information of the reference information.

[0163] The terminal device 40 receives the generated results from the report device 10. The terminal device 40 displays a display screen 640 on a display device. A report included in the generated results is displayed on the display screen 640. Auxiliary information included in the report may also be displayed on the display screen 640.

[0164] In step S9, the terminal device 40 determines whether or not to reselect the reference information. Specifically, when the reselect button 644 on the display screen 640 is pressed, the terminal device 40 determines that the reference information is to be reselected. On the other hand, when the reselect button 644 on the display screen 640 is not pressed, the terminal device 40 determines that the reference information is not to be reselected.

[0165] If the reference information is to be reselected (YES), the terminal device 40 displays the selection screen 630 again. The terminal device 40 transmits a selection request to the report device 10 in response to a user's operation on the selection screen 630. When the report device 10 receives the selection request from the terminal device 40, it returns the process to step S5. The terminal device 40 returns the process to step S2 and may display the condition setting screen 620 to change the search conditions and search again for the reference conditions. On the other hand, if the reference information is not to be reselected (NO), the terminal device 40 proceeds to step S10.

[0166] In step S10, it is determined whether or not to regenerate the report. Specifically, when the regeneration button 643 on the display screen 640 is pressed, the terminal device 40 determines that the report is to be regenerated. On the other hand, when the regeneration button 643 on the display screen 640 is not pressed, the terminal device 40 determines that the report is not to be regenerated.

[0167] If the report is to be regenerated (YES), the terminal device 40 displays the format designation screen 600 again. The terminal device 40 transmits a generation request to the report device 10 in response to a user's operation on the format designation screen 600. When the report device 10 receives the generation request from the terminal device 40, it returns the process to step S1. On the other hand, if the report is not to be regenerated (NO), the terminal device 40 ends the report generation process.

[0168] <Modification> The report generation process may be performed in the following manner.

[0169] (1) The user specifies information about the output format. The user may select one output format from a plurality of output formats. The output format may be determined in advance. The user may specify information about the output format through text editing or the like.

[0170] (2) The user specifies one or more pieces of reference information to be used in generating output information. In this modified example, the search for reference information by the search unit 120 may be omitted. The user may specify one or more pieces of reference information through a drag-and-drop operation or the like. The method for specifying one or more pieces of reference information by the user may be the method described in the input acceptance of new reference information.

[0171] (3) The generation unit 150 generates input information for the machine learning model based on the information on the output format and one or more pieces of reference information specified by the user. The generation unit 150 generates output information by inputting the generated input information to the machine learning model, and presents the output information to the user.

[0172] (4) The output unit 170 returns to the process of (1) or (2) and generates different output information. If the user wishes to further improve the output information, the user may return to the process of (1) or (2). The output unit 170 may generate difference information between multiple reports and present it to the user. (1) to (4) may be executed repeatedly until specified by the user or until a predetermined number of times has been reached.

[0173] (5) Finally, the output information selected by the user is output in a predetermined output format (e.g., PDF file, etc.). The output format may be specified by the user in (1).

[0174] [Second embodiment] In the first embodiment, a configuration is described in which output format information for all items included in a report is included in the model input information, and output information showing the entire report is generated in a foundation model. In the second embodiment, a configuration is described in which output format information for each item included in a report is included in the model input information, output information for each item is generated based on the foundation model, and output information showing the entire report is generated by integrating the output information for each item.

[0175] <Report generation process flow> The report generation process according to this embodiment will be described with reference to Fig. 10. Fig. 10 is a flowchart showing an example of the report generation process according to the second embodiment. The report generation process according to this embodiment will be described below, focusing on the differences from the first embodiment.

[0176] In step S11, the request receiving unit 110 of the report device 10 receives a generation request from the terminal device 40. The output format information and user input information included in the generation request include information about a plurality of items to be included in the report.

[0177] The generation request may include search conditions. The search conditions may include at least one of search conditions for the entire report or search conditions for each item to be included in the report. The search conditions included in the generation request may be set by an operation on the condition setting screen 620.

[0178] In step S12, the search unit 120 of the report device 10 selects one item from among the items included in the report. The items included in the report may be in a pre-determined state or may not be pre-determined. When the items included in the report are not pre-determined, the user may specify the contents of the headings, the order of the headings, or the hierarchical information of the headings by text input.

[0179] The search unit 120 may select any item included in the report. As an example, the search unit 120 may select the first item in the report. The first item in the report may be, for example, the title of the report. The search unit 120 obtains user input information and search conditions related to the selected item from the generation request.

[0180] The search unit 120 transmits a search request to the search device 20 based on the acquired user input information and search conditions. When the search conditions acquired from the generation request include both search conditions related to the entire report and search conditions related to items, the search unit 120 may generate a search request based on the search conditions related to the items. In this case, the search unit 120 may generate a search request based on each of the search conditions related to the entire report and the search conditions related to the items. When a search is performed based on each of the search conditions related to the entire report and the search conditions related to the items, the list display unit 631 of the selection screen 630 may display information indicating which search conditions each piece of reference information was searched based on.

[0181] The search device 20 receives a search request from the report device 10. The search device 20 searches for reference information from the data source 25 indicated in the search conditions based on the search request. The search device 20 transmits the search results for the reference information to the report device 10. The search unit 120 of the report device 10 receives the search results for the reference information from the search device 20. The search unit 120 transmits the search results for the reference information to the presentation unit 130.

[0182] The processes from step S13 to step S15 are the same as those from step S3 to step S5 in the first embodiment. Note that, like the first embodiment, the processes from step S13 to step S15 may be omitted.

[0183] In step S16, the generation unit 150 of the report device 10 generates model input information based on the output format information and user input information related to the items acquired in step S12, and the reference information acquired in step S15 (or the reference information included in the search result acquired in step S12). That is, the generation unit 150 generates model input information that instructs the generation of output information for each item included in the report. The generation unit 150 transmits the generated model input information to the generation device 30.

[0184] In step S17, the generating device 30 inputs the model input information received from the reporting device 10 to the base model 35. The base model 35 executes a predetermined task based on the model input information, and outputs the output information generated as a result. The generating device 30 transmits the output information output from the base model 35 to the reporting device 10.

[0185] The generating device 30 may generate the output information by inputting the model input information to a different base model for each item. The base model to be input for each item may be determined based on an instruction from a user.

[0186] The generation unit 150 of the report device 10 receives output information from the generation device 30. The output information received here is output information corresponding to the item selected in step S12. The report device 10 returns the process to step S12. When the process returns to step S12, the search unit 120 selects any item included in the report for which output information has not been generated. As an example, the search unit 120 may select items for which output information has not been generated, in order from the top of the report. The search unit 120 may select an item specified by the user.

[0187] After that, the report device 10 executes the processes from step S12 to step S17 again. In this way, the report device 10 repeatedly executes the search for reference information and the generation of output information until the output information is acquired for all items included in the report. When the generation unit 150 acquires the output information for all items included in the report, it sends the output information for each item acquired in step S17 to the output unit 170.

[0188] In step S18, the output unit 170 of the report device 10 receives the output information for each item from the generation unit 150. The output unit 170 generates a report based on the output information for each item. The output unit 170 transmits the generation result including the generated report to the terminal device 40. The report may include auxiliary information of the reference information used to generate the output information for each item.

[0189] The output unit 170 may generate a report by integrating the output information for each item. As an example, the output unit 170 may generate a report by combining the output information for each item in the order acquired by the generation unit 150. The output unit 170 may also generate a report by combining the output information for each item in the order of the items indicated in the output format information.

[0190] As an example, the output unit 170 may integrate the output information for each item based on the foundation model 35. Specifically, the output unit 170 may generate model input information based on the configuration information of the items included in the report and the output information for each item, and cause the foundation model 35 to generate a report based on the model input information. The configuration information is information indicating the configuration of the items included in the report. As an example, the configuration information may be information listing the titles and headings included in the report. The configuration information may be output format information, or may be information obtained by processing the output format information. The output unit 170 may integrate the output information for each item based on another foundation model. The other foundation model may be a foundation model used to generate the output information for each item, or may be a foundation model different from the foundation model.

[0191] The output unit 170 may generate model input information including configuration information of items included in the report, a report combining output information for each item, and instruction information for instructing formatting of the report, and may cause the base model 35 to generate the report based on the model input information. The instruction information may include constraint information for instructing to maintain the output information for each item.

[0192] All configurations related to the information processing system 1000 according to the first embodiment are applicable to the information processing system 1000 according to the second embodiment. Similarly, all configurations related to the information processing system 1000 according to the second embodiment are applicable to the information processing system 1000 according to the first embodiment.

[0193] <Summary> As is clear from the above explanation, an information processing system according to one embodiment of the present disclosure acquires information regarding an output format, searches for multiple reference information based on a search request, acquires one or more reference information to be used for generating output information based on the search results, generates input information for a machine learning model based on the information regarding the output format and the one or more reference information, and generates the output information by inputting the input information to the machine learning model.

[0194] The information processing system may output information about the search results and obtain one or more pieces of reference information based on a selection by a user. The information about the search results may include at least one of the data source, link information, at least a portion of the information, a summary, a thumbnail image, an owner, a copyright holder, a creation date, or terms of use of the reference information. The information processing system may obtain reference information other than the plurality of reference information as one or more pieces of reference information to be used in generating output information based on an instruction from a user.

[0195] The information on the output format may include information on one or more output formats selected from a plurality of output formats. The information on the output format may include information on a structure of the output information. The information on the structure may include information on one or more items to be included in the output information. The information on the one or more items may include information on the names of the items and the hierarchy of the items to be included in the output information.

[0196] The search request may include at least one of search conditions or a query generated based on user input information. The search conditions may include at least one of the data source, type, creation date, or usage conditions to be searched. The information processing system may set the search conditions based on user information. The user information may include at least one of the user's authority, attribute, affiliation, job title, or usage plan.

[0197] The information processing system may obtain one or more other reference information to be used in generating the second output information based on the search result, generate second input information based on the information on the output format and the one or more other reference information, and generate the second output information by inputting the second input information to the machine learning model. The information processing system may output difference information between the output information and the second output information.

[0198] The information processing system may search for other multiple reference information based on another search request, acquire one or more other reference information to be used for generating second output information based on a result of the search for the other multiple reference information, generate second input information for the machine learning model based on the information on the output format and the other one or more reference information, and generate the second output information by inputting the second input information to the machine learning model. The information processing system may output difference information between the output information and the second output information.

[0199] The information processing system may output, together with the output information, at least one of the data source, at least a portion of the information, a summary, a thumbnail image, an owner, a copyright holder, a creation date, or terms of use of the one or more reference information. The information processing system may output additional information regarding the output information before generating the output information. The additional information may include at least one of the time, cost, or computing resources required to generate the output information.

[0200] The information processing system may output the usage conditions. The usage conditions may include information indicating whether or not the use of the reference information is chargeable. The information processing system may determine a cost required for generating or using the output information based on at least information indicating whether or not each of the one or more pieces of reference information is chargeable. The information processing system may output the determined cost before executing a generation process of the output information. After outputting the determined cost, if the information processing system receives a generation instruction from a user, it may generate the output information by inputting the input information into a machine learning model.

[0201] As a result, according to an embodiment of the present disclosure, a technique capable of acquiring desired output information can be provided. According to one aspect, according to an embodiment, a report written in a desired output format can be easily generated.

[0202] [Hardware configuration of information processing device] A part or all of each device (report device 10, search device 20, generation device 30, and terminal device 40) in the above-mentioned embodiment may be configured with hardware, or may be configured with information processing of software (programs) executed by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like. In the case of being configured with information processing of software, software for realizing at least a part of the functions of each device in the above-mentioned embodiment may be stored in a non-transient storage medium (non-transient computer-readable medium) such as a CD-ROM (Compact Disc-Read Only Memory) or a USB (Universal Serial Bus) memory, and the information processing of the software may be executed by reading the software into a computer. The software may also be downloaded via a communication network. Furthermore, the information processing by the software may be executed by hardware by implementing all or a part of the processing of the software in a circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0203] The storage medium that stores the software may be a removable medium such as an optical disk, or a fixed storage medium such as a hard disk, memory, etc. The storage medium may be provided inside the computer (main storage device, auxiliary storage device, etc.) or outside the computer.

[0204] 11 is a block diagram showing an example of a hardware configuration of each device (report device 10, search device 20, generation device 30, and terminal device 40) in the above-mentioned embodiment. Each device may be realized as a computer 7 including, for example, a processor 71, a main storage device 72 (memory), an auxiliary storage device 73 (memory), a network interface 74, and a device interface 75, which are connected via a bus 76.

[0205] Although the computer 7 in FIG. 11 includes one of each component, it may include a plurality of the same components. Although one computer 7 is shown in FIG. 11, the software may be installed in a plurality of computers, and each of the plurality of computers may execute the same or different parts of the software. In this case, it may be a form of distributed computing in which each computer communicates via a network interface 74 or the like to execute the processing. That is, each device in the above-mentioned embodiment (the report device 10, the search device 20, the generation device 30, and the terminal device 40) may be configured as a system in which one or more computers execute instructions stored in one or more storage devices to realize the function. Also, it may be configured such that information transmitted from the terminal is processed by one or more computers provided on the cloud, and the processing results are transmitted to the terminal.

[0206] Various calculations of each device (report device 10, search device 20, generation device 30, and terminal device 40) in the above-mentioned embodiment may be executed in parallel using one or more processors, or using multiple computers via a network. Also, various calculations may be distributed to multiple computing cores in a processor and executed in parallel. Also, some or all of the processes, means, etc. disclosed herein may be realized by at least one of a processor and a storage device provided on a cloud that can communicate with computer 7 via a network. In this way, each device in the above-mentioned embodiment may be in the form of parallel computing using one or more computers.

[0207] The processor 71 may be an electronic circuit (processing circuit, processing circuitry, CPU, GPU, FPGA, ASIC, etc.) that at least controls a computer or performs calculations. The processor 71 may be any of a general-purpose processor, a dedicated processing circuit designed to perform a specific calculation, or a semiconductor device including both a general-purpose processor and a dedicated processing circuit. The processor 71 may also include an optical circuit, or may include a calculation function based on quantum computing.

[0208] The processor 71 may perform arithmetic processing based on data or software input from each device or the like configured inside the computer 7, and may output arithmetic results or control signals to each device or the like. The processor 71 may control each component constituting the computer 7 by executing the OS (Operating System) of the computer 7, applications, or the like.

[0209] Each of the devices in the above-described embodiment (report device 10, search device 20, generation device 30, and terminal device 40) may be realized by one or more processors 71. Here, the processor 71 may refer to one or more electronic circuits arranged on one chip, or to one or more electronic circuits arranged on two or more chips or two or more devices. When multiple electronic circuits are used, the electronic circuits may communicate with each other by wire or wirelessly.

[0210] The main memory 72 may store instructions executed by the processor 71 and various data, and information stored in the main memory 72 may be read by the processor 71. The auxiliary memory 73 is a memory device other than the main memory 72. These memory devices refer to any electronic components capable of storing electronic information, and may be semiconductor memories. The semiconductor memories may be either volatile or non-volatile memories. The memory devices for saving various data in each device (report device 10, search device 20, generation device 30, and terminal device 40) in the above-mentioned embodiment may be realized by the main memory 72 or the auxiliary memory 73, or may be realized by an internal memory built into the processor 71. For example, each memory unit in the above-mentioned embodiment may be realized by the main memory 72 or the auxiliary memory 73.

[0211] When each device (report device 10, search device 20, generation device 30, and terminal device 40) in the above-mentioned embodiment is composed of at least one storage device (memory) and at least one processor connected (coupled) to this at least one storage device, at least one processor may be connected to one storage device. Also, at least one storage device may be connected to one processor. Also, a configuration may be included in which at least one processor of the multiple processors is connected to at least one storage device of the multiple storage devices. Also, this configuration may be realized by storage devices and processors included in multiple computers. Furthermore, a configuration may be included in which a storage device is integrated with a processor (for example, a cache memory including an L1 cache and an L2 cache).

[0212] The network interface 74 is an interface for connecting to the communication network 8 wirelessly or by wire. The network interface 74 may be an appropriate interface, such as one conforming to an existing communication standard. The network interface 74 may exchange information with an external device 9A connected via the communication network 8. The communication network 8 may be any one of a wide area network (WAN), a local area network (LAN), a personal area network (PAN), etc., or a combination thereof, as long as information is exchanged between the computer 7 and the external device 9A. An example of a WAN is the Internet, an example of a LAN is IEEE802.11 or Ethernet (registered trademark), and an example of a PAN is Bluetooth (registered trademark) or Near Field Communication (NFC).

[0213] The device interface 75 is an interface such as a USB that directly connects to the external device 9B.

[0214] The external device 9A is a device connected to the computer 7 via a network. The external device 9B is a device connected directly to the computer 7.

[0215] The external device 9A or the external device 9B may be, for example, an input device. The input device is, for example, a device such as a camera, a microphone, a motion capture device, various sensors, a keyboard, a mouse, a touch panel, or the like, and provides acquired information to the computer 7. Alternatively, the input device may be a device equipped with an input unit, a memory, and a processor, such as a personal computer, a tablet terminal, or a smartphone.

[0216] Moreover, the external device 9A or the external device 9B may be, for example, an output device. The output device may be, for example, a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) panel, or a speaker that outputs sound or the like. Alternatively, the output device may be a device including an output unit, a memory, and a processor, such as a personal computer, a tablet terminal, or a smartphone.

[0217] Furthermore, the external device 9A or the external device 9B may be a storage device (memory). For example, the external device 9A may be a network storage or the like, and the external device 9B may be a storage device such as an HDD.

[0218] Furthermore, the external device 9A or the external device 9B may be a device having some of the functions of the components of each device (the report device 10, the search device 20, the generation device 30, and the terminal device 40) in the above-mentioned embodiment. In other words, the computer 7 may transmit some or all of the processing results to the external device 9A or the external device 9B, or may receive some or all of the processing results from the external device 9A or the external device 9B.

[0219] In this specification (including the claims), when the expression "at least one of a, b, and c" or "at least one of a, b, or c" (including similar expressions) is used, it includes any of a, b, c, ab, ac, bc, or abc. It may also include multiple instances of any element, such as aa, abb, aabbcc, etc. Furthermore, it also includes the addition of elements other than the enumerated elements (a, b, and c), such as having d, as in abcd.

[0220] In this specification (including claims), when expressions such as "using data as input / based on / according to / in response to data" (including similar expressions) are used, unless otherwise specified, this includes the use of data itself or data that has been processed in some way (e.g., data with noise added, normalized, feature values ​​extracted from data, intermediate representations of data, etc.). In addition, when it is stated that a result is obtained "using data as input / based on / according to / in response to data" (including similar expressions), unless otherwise specified, this includes the use of the result based only on the data, or the use of the result influenced by other data, factors, conditions, and / or states other than the data. In addition, when it is stated that "data is output" (including similar expressions), unless otherwise specified, this includes the use of data itself as output, or data that has been processed in some way (e.g., data with noise added, normalized, feature values ​​extracted from data, intermediate representations of various data, etc.) as output.

[0221] In this specification (including the claims), the terms "connected" and "coupled" are intended as open-ended terms including any of direct connection / coupling, indirect connection / coupling, electrically connection / coupling, communicatively connection / coupling, functionally connection / coupling, physically connection / coupling, etc. The terms should be interpreted appropriately depending on the context in which the terms are used, but any connection / coupling form that is not intentionally or naturally excluded should be interpreted as being included in the terms without any restriction.

[0222] In this specification (including the claims), when the expression "A configured to B" is used, it may include that the physical structure of element A has a configuration capable of performing operation B, and that the permanent or temporary setting / configuration of element A is configured / set to actually perform operation B. For example, when element A is a general-purpose processor, it is sufficient that the processor has a hardware configuration capable of performing operation B, and is configured to actually perform operation B by setting a permanent or temporary program (instruction). Also, when element A is a dedicated processor, dedicated arithmetic circuit, etc., it is sufficient that the circuit structure, etc. of the processor is implemented to actually perform operation B, regardless of whether control instructions and data are actually attached.

[0223] In this specification (including the claims), when terms implying containing or possessing (e.g., "comprising / including," "having," etc.) are used, they are intended to be open-ended terms that include cases in which the term contains or possesses things other than the object indicated by the object of the term. When the object of such terms implying containing or possessing does not specify a quantity or suggests a singular number (such as an expression using the article "a" or "an"), the expression should be construed as not being limited to a specific number.

[0224] In this specification (including the claims), even if expressions such as "one or more" and "at least one" are used in some places and expressions that do not specify a quantity or suggest a singular number (expressions using the articles "a" or "an") are used in other places, the latter expressions are not intended to mean "one." In general, expressions that do not specify a quantity or suggest a singular number (expressions using the articles "a" or "an") should be interpreted as not necessarily being limited to a specific number.

[0225] In this specification, when a particular advantage / result is described as being obtained from a particular configuration of an embodiment, it should be understood that the same effect can also be obtained from one or more other embodiments having the same configuration, unless there is a reason to the contrary. However, it should be understood that the presence or absence of the effect generally depends on various factors, conditions, and / or states, and that the effect is not necessarily obtained by the configuration. The effect is merely obtained by the configuration described in the embodiment when various factors, conditions, and / or states are satisfied, and the effect is not necessarily obtained in the invention according to the claim that specifies the configuration or a similar configuration.

[0226] In this specification (including claims), when a plurality of pieces of hardware perform a predetermined process, each piece of hardware may cooperate to perform the predetermined process, or a portion of the hardware may perform all of the predetermined process. Also, a portion of the hardware may perform a portion of the predetermined process, and another piece of hardware may perform the remainder of the predetermined process. In this specification (including claims), when an expression such as "one or more pieces of hardware perform a first process, and the one or more pieces of hardware perform a second process" (including similar expressions) is used, the hardware performing the first process and the hardware performing the second process may be the same or different. In other words, it is sufficient that the hardware performing the first process and the hardware performing the second process are included in the one or more pieces of hardware. The hardware may include an electronic circuit, a device including an electronic circuit, and the like.

[0227] In this specification (including the claims), when multiple storage devices (memories) store data, each of the multiple storage devices may store only a portion of the data, or may store the entire data. Also, a configuration in which only some of the multiple storage devices store data may be included.

[0228] In this specification (including the claims), terms such as "first", "second", etc. are used merely as a way of distinguishing between two or more elements, and are not necessarily intended to impose technical meanings such as time aspect, space aspect, sequence, quantity, etc. on the subject matter. Thus, for example, a reference to a first element and a second element does not necessarily mean that only two elements may be employed therein, that the first element must precede the second element, that the first element must be present for the second element to be present, etc.

[0229] Although the embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to the above-mentioned individual embodiments. Various additions, modifications, replacements, partial deletions, etc. are possible within the scope of the conceptual idea and intent of the present invention derived from the contents defined in the claims and their equivalents. For example, in the above-mentioned embodiments, when numerical values ​​or formulas are used for explanation, these are shown for illustrative purposes and do not limit the scope of the present disclosure. In addition, the order of each operation shown in the embodiment is also illustrative and does not limit the scope of the present disclosure.

[0230] In addition, the disclosed technology may take the following additional features.

[0231] (Appendix 1) At least one memory; at least one processor; The at least one processor: Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; The input information is input to the machine learning model to generate the output information. Information processing system.

[0232] (Appendix 2) The at least one processor: outputting information regarding the results of said search; Obtaining said one or more references based on a selection made by a user; 2. An information processing system according to claim 1.

[0233] (Appendix 3) The information about the search results includes at least one of the data source, link information, at least a part of the information, a summary, a thumbnail image, an owner, a copyright holder, a creation date, or a condition of use of the reference information; 3. An information processing system according to claim 2.

[0234] (Appendix 4) The at least one processor: acquiring reference information other than the plurality of pieces of reference information as the one or more pieces of reference information to be used in generating the output information based on an instruction from the user; 4. An information processing system according to claim 3.

[0235] (Appendix 5) The information regarding the output format includes information regarding one or more output formats selected from a plurality of output formats. 5. An information processing system according to any one of claims 1 to 4.

[0236] (Appendix 6) The information regarding the output format includes information regarding a structure of the output information. 6. An information processing system according to claim 5.

[0237] (Appendix 7) The information regarding the structure includes information regarding one or more items to be included in the output information. 7. An information processing system according to claim 6.

[0238] (Appendix 8) The information about the one or more items includes information about the item names and hierarchical structure of the items to be included in the output information. 8. An information processing system according to claim 7.

[0239] (Appendix 9) The search request includes at least one of a search criterion or a query generated based on user input information; 9. An information processing system according to any one of appendix 1 to 8.

[0240] (Appendix 10) The search conditions include at least one of the data source, type, creation date, or usage conditions to be searched; 10. The information processing system according to claim 9.

[0241] (Appendix 11) The at least one processor: setting the search conditions based on user information; 11. The information processing system according to claim 9 or 10.

[0242] (Appendix 12) The user information includes at least one of the user's authority, attribute, affiliation, position, or usage plan. 12. The information processing system according to claim 11.

[0243] (Appendix 13) The at least one processor: obtaining one or more other pieces of reference information for use in generating second output information based on a result of the search; generating second input information based on the information on the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; 13. An information processing system according to any one of appendix 1 to 12.

[0244] (Appendix 14) The at least one processor: outputting difference information between the output information and the second output information; 14. The information processing system according to claim 13.

[0245] (Appendix 15) The at least one processor: Retrieving other references based on other search requests; obtaining one or more other pieces of reference information to be used in generating second output information based on a result of searching the other plurality of pieces of reference information; generating second input information for the machine learning model based on the information about the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; 15. An information processing system according to any one of claims 1 to 14.

[0246] (Appendix 16) The at least one processor: outputting difference information between the output information and the second output information; 16. An information processing system according to claim 15.

[0247] (Appendix 17) The at least one processor: outputting at least one of a data source, at least a part of information, a summary, a thumbnail image, an owner, a copyright holder, a creation date, or a condition of use of the one or more reference information together with the output information; 17. An information processing system according to any one of claims 1 to 16.

[0248] (Appendix 18) The at least one processor: outputting additional information related to the output information before generating the output information; 18. An information processing system according to any one of appendix 1 to 17.

[0249] (Appendix 19) the additional information includes at least one of the time, cost, or computational resources required to generate the output information; 19. The information processing system according to claim 18.

[0250] (Appendix 20) The at least one processor: Outputting the terms of use; the use conditions include information indicating whether or not use of the reference information is chargeable; 5. An information processing system according to claim 3 or 4.

[0251] (Appendix 21) The at least one processor: determining a cost required for generating or using the output information based on at least information indicating whether each of the one or more pieces of reference information is chargeable; 21. The information processing system according to claim 20.

[0252] (Appendix 22) The at least one processor: outputting the determined cost before executing the process of generating the output information; 22. The information processing system according to claim 21.

[0253] (Appendix 23) The at least one processor: and after outputting the determined cost, if a generation instruction is received from a user, generating the output information by inputting the input information into the machine learning model. 23. An information processing system according to claim 22.

[0254] (Appendix 24) At least one memory; at least one processor; The at least one processor: Displaying information for specifying the output format on a display device, displaying a result of searching the plurality of pieces of reference information based on the search request on the display device; displaying information for selecting one or more pieces of reference information from the plurality of pieces of reference information on the display device; displaying, on the display device, output information generated by inputting input information to a machine learning model generated based on information regarding the specified output format and the one or more selected reference information to the machine learning model; Information processing device.

[0255] (Appendix 25) The at least one processor: displaying, as a result of searching the plurality of reference information, at least one of a data source, link information, at least a part of information, a summary, a thumbnail image, an owner, a copyright holder, a creation date, and a usage condition of the reference information on the display device; 25. The information processing device according to claim 24.

[0256] (Appendix 26) The at least one processor: displaying information for designating reference information other than the plurality of reference information on the display device; 26. The information processing device according to claim 25.

[0257] (Appendix 27) The at least one processor: displaying information for selecting the output format from a plurality of output formats on the display device; 27. An information processing device according to any one of appendix 24 to 26.

[0258] (Appendix 28) The at least one processor: displaying, together with the output information, at least one of a data source, at least a part of information, a summary, a thumbnail image, an owner, a copyright holder, a creation date, or a usage condition of the one or more reference information on the display device; 28. An information processing device according to any one of appendix 24 to 27.

[0259] (Appendix 29) The at least one processor: displaying, on the display device, information indicating whether or not each of the plurality of pieces of reference information has been selected, as information for selecting one or more pieces of reference information from the plurality of pieces of reference information; 29. An information processing device according to any one of appendix 24 to 28.

[0260] (Appendix 30) The at least one processor: displaying information related to drag and drop on the display device as information for designating reference information other than the plurality of reference information; 27. The information processing device according to claim 26.

[0261] (Appendix 31) At least one memory; at least one processor; The at least one processor: Obtaining a first search request for a first item; retrieving a plurality of first reference information based on the first search request; obtaining one or more pieces of first reference information to be used in generating first output information based on a result of the search based on the first search request; generating first input information for a machine learning model based on the information about the first item and the one or more first reference information; generating the first output information by inputting the first input information into the machine learning model; Obtain a second search request for a second item; retrieving a plurality of second reference information based on the second search request; obtaining one or more pieces of second reference information to be used in generating second output information based on a result of the search based on the second search request; generating second input information for the machine learning model based on information about a second item and the one or more second reference information; generating the second output information by inputting the second input information into the machine learning model; generating third output information based on the first output information and the second output information; Information processing system.

[0262] (Appendix 32) The at least one processor: outputting information regarding a result of the search based on the first search request; Obtaining the one or more first references based on a selection by a user; outputting information regarding a result of the search based on the second search request; obtaining the one or more second reference information based on a selection by the user; 32. The information processing system of claim 31.

[0263] (Appendix 33) The at least one processor: generating third input information for the machine learning model based on the first output information and the second output information; generating the third output information by inputting the third input information into the machine learning model; 33. The information processing system according to claim 31 or 32.

[0264] (Appendix 34) At least one processor Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; The input information is input to the machine learning model to generate the output information. An information processing method for performing a process.

[0265] (Appendix 35) At least one processor Displaying information for specifying the output format on a display device, displaying a result of searching the plurality of pieces of reference information based on the search request on the display device; displaying information for selecting one or more pieces of reference information from the plurality of pieces of reference information on the display device; displaying, on the display device, output information generated by inputting input information to a machine learning model generated based on information regarding the specified output format and the one or more selected reference information to the machine learning model; An information processing method for performing a process.

[0266] (Appendix 36) At least one processor Obtaining a first search request for a first item; retrieving a plurality of first reference information based on the first search request; obtaining one or more pieces of first reference information to be used in generating first output information based on a result of the search based on the first search request; generating first input information for a machine learning model based on the information about the first item and the one or more first reference information; generating the first output information by inputting the first input information into the machine learning model; Obtain a second search request for a second item; retrieving a plurality of second reference information based on the second search request; obtaining one or more pieces of second reference information to be used in generating second output information based on a result of the search based on the second search request; generating second input information for the machine learning model based on information about a second item and the one or more second reference information; generating the second output information by inputting the second input information into the machine learning model; generating third output information based on the first output information and the second output information; An information processing method for performing a process.

[0267] (Appendix 37) At least one processor has Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; The input information is input to the machine learning model to generate the output information. A program for executing a process.

[0268] (Appendix 38) At least one processor has Displaying information for specifying the output format on a display device, displaying a result of searching the plurality of pieces of reference information based on the search request on the display device; displaying information for selecting one or more pieces of reference information from the plurality of pieces of reference information on the display device; displaying, on the display device, output information generated by inputting input information to a machine learning model generated based on information regarding the specified output format and the one or more selected reference information to the machine learning model; A program for executing a process.

[0269] (Appendix 39) At least one processor has Obtaining a first search request for a first item; retrieving a plurality of first reference information based on the first search request; obtaining one or more pieces of first reference information to be used in generating first output information based on a result of the search based on the first search request; generating first input information for a machine learning model based on the information about the first item and the one or more first reference information; generating the first output information by inputting the first input information into the machine learning model; Obtain a second search request for a second item; retrieving a plurality of second reference information based on the second search request; obtaining one or more pieces of second reference information to be used in generating second output information based on a result of the search based on the second search request; generating second input information for the machine learning model based on information about a second item and the one or more second reference information; generating the second output information by inputting the second input information into the machine learning model; generating third output information based on the first output information and the second output information; A program for executing a process. [Explanation of symbols]

[0270] 10: Reporting device 20: Search device 25: Data source 30:Generation device 35: Base model 40: Terminal device 110: Request reception unit 120: Search section 130: Presentation part 140: Selection reception section 150: Generation part 160: History memory section 170: Output section 180: Billing Department 1000: Information Processing Systems

Claims

1. At least one memory; At least one processor; The at least one processor Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; generating the output information by inputting the input information into the machine learning model; obtaining one or more other pieces of reference information for use in generating second output information based on a result of the search; generating second input information for the machine learning model based on the information related to the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; outputting difference information between the output information and the second output information; Information processing system.

2. At least one memory; At least one processor; The at least one processor Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; generating the output information by inputting the input information into the machine learning model; Retrieving other references based on other search requests; obtaining one or more pieces of other reference information to be used in generating second output information based on a result of searching the other plurality of pieces of reference information; generating second input information for the machine learning model based on the information related to the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; outputting difference information between the output information and the second output information; Information processing system.

3. The at least one processor presenting information to the user regarding the results of said search; receiving a selection request for reference information for the presentation from the user; obtaining the one or more pieces of reference information based on the selection request; 3. The information processing system according to claim 1 or 2.

4. The at least one processor presenting to the user, as information related to the search results, at least one of the following: a data source of the reference information, link information, at least a part of the information, a summary, a thumbnail image, a title, an outline, an owner, a copyright holder, a creation date, or terms of use; The information processing system according to claim 3 .

5. The at least one processor presenting to the user one or more output formats containing information regarding the structure of the report; obtaining information regarding the output format based on the user's selection of the presentation; 3. The information processing system according to claim 1 or 2.

6. The information regarding the output format obtained includes information input or edited by the user.

6. The information processing system according to claim 5.

7. The information regarding the structure of the report includes information regarding the names of items to be included in the output information, the order of items, and the hierarchy of items.

6. The information processing system according to claim 5.

8. The at least one processor setting search conditions included in the search request based on user information; The user information includes at least one of the user's authority, attribute, affiliation, position, or usage plan.

3. The information processing system according to claim 1 or 2.

9. The at least one processor outputting at least one of a data source, at least a part of information, a summary, a thumbnail image, a title, a summary, an owner, a copyright holder, a creation date, or a usage condition of the one or more reference information together with the output information; 3. The information processing system according to claim 1 or 2.

10. The at least one processor outputting additional information related to the output information before generating the output information; the additional information includes at least one of the time, cost, or computational resources required to generate the output information; 3. The information processing system according to claim 1 or 2.

11. The information regarding the output format includes information regarding a structure of the report; The input information includes information regarding the structure of the report; The at least one processor searching the plurality of references using a query based on information regarding the structure of the report; 3. The information processing system according to claim 1 or 2.

12. At least one memory; At least one processor; The at least one processor Obtaining a first search request for a first item; retrieving a plurality of first reference information based on the first search request; obtaining one or more pieces of first reference information to be used in generating first output information based on a result of the search based on the first search request; generating first input information for a machine learning model based on information about the first item and the one or more first reference information; inputting the first input information into the machine learning model to generate the first output information; Obtaining a second search request for a second item; retrieving a plurality of second reference information based on the second search request; obtaining one or more pieces of second reference information to be used in generating second output information based on a result of the search based on the second search request; generating second input information for the machine learning model based on the information about the second item and the one or more second reference information; generating the second output information by inputting the second input information into the machine learning model; generating third input information for the machine learning model based on the first output information and the second output information; generating third output information by inputting the third input information into the machine learning model; Information processing system.

13. The third input information includes configuration information of the first item and the second item. The information processing system according to claim 12.

14. The at least one processor outputting information regarding a result of the search based on the first search request; obtaining the one or more first references based on a selection by a user; outputting information regarding a result of the search based on the second search request; obtaining the one or more second reference information based on a selection by the user; 14. An information processing system according to claim 12 or 13.

15. The information about the result of the search based on the first search request includes at least one of a data source, link information, at least a part of information, a summary, a thumbnail image, a title, a summary, an owner, a copyright holder, a creation date, or a usage condition of the first reference information; The information about the result of the search based on the second search request includes at least one of a data source, link information, at least a part of information, a summary, a thumbnail image, a title, a summary, an owner, a copyright holder, a creation date, or a usage condition of the second reference information.

15. The information processing system according to claim 14.

16. The at least one processor setting search conditions to be included in the first search request based on user information; The user information includes at least one of the user's authority, attribute, affiliation, position, or usage plan.

14. An information processing system according to claim 12 or 13.

17. The at least one processor outputting, together with the third output information, at least one of a data source, at least a part of information, a summary, a thumbnail image, a title, a summary, an owner, a copyright holder, a creation date, or a usage condition of the one or more first reference information, and at least one of a data source, at least a part of information, a summary, a thumbnail image, a title, a summary, an owner, a copyright holder, a creation date, or a usage condition of the one or more second reference information; 14. An information processing system according to claim 12 or 13.

18. At least one processor Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; generating the output information by inputting the input information into the machine learning model; obtaining one or more other pieces of reference information for use in generating second output information based on a result of the search; generating second input information for the machine learning model based on the information related to the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; outputting difference information between the output information and the second output information; An information processing method for performing a process.

19. At least one processor Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; generating the output information by inputting the input information into the machine learning model; Retrieving other references based on other search requests; obtaining one or more pieces of other reference information to be used in generating second output information based on a result of searching the other plurality of pieces of reference information; generating second input information for the machine learning model based on the information related to the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; outputting difference information between the output information and the second output information; An information processing method for performing a process.

20. At least one processor Obtaining a first search request for a first item; retrieving a plurality of first reference information based on the first search request; obtaining one or more pieces of first reference information to be used in generating first output information based on a result of the search based on the first search request; generating first input information for a machine learning model based on information about the first item and the one or more first reference information; inputting the first input information into the machine learning model to generate the first output information; Obtaining a second search request for a second item; retrieving a plurality of second reference information based on the second search request; obtaining one or more pieces of second reference information to be used in generating second output information based on a result of the search based on the second search request; generating second input information for the machine learning model based on the information about the second item and the one or more second reference information; generating the second output information by inputting the second input information into the machine learning model; generating third input information for the machine learning model based on the first output information and the second output information; generating third output information by inputting the third input information into the machine learning model; An information processing method for performing a process.

21. At least one processor, Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; generating the output information by inputting the input information into the machine learning model; obtaining one or more other pieces of reference information for use in generating second output information based on a result of the search; generating second input information for the machine learning model based on the information related to the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; outputting difference information between the output information and the second output information; A program for executing a process.

22. At least one processor, Get information about the output format, Retrieving a plurality of references based on a search request; obtaining one or more pieces of reference information for use in generating output information based on a result of the search; generating input information for a machine learning model based on the information regarding the output format and the one or more reference information; generating the output information by inputting the input information into the machine learning model; Retrieving other references based on other search requests; obtaining one or more pieces of other reference information to be used in generating second output information based on a result of searching the other plurality of pieces of reference information; generating second input information for the machine learning model based on the information related to the output format and the one or more other reference information; generating the second output information by inputting the second input information into the machine learning model; outputting difference information between the output information and the second output information; A program for executing a process.

23. At least one processor, Obtaining a first search request for a first item; retrieving a plurality of first reference information based on the first search request; obtaining one or more pieces of first reference information to be used in generating first output information based on a result of the search based on the first search request; generating first input information for a machine learning model based on information about the first item and the one or more first reference information; inputting the first input information into the machine learning model to generate the first output information; Obtaining a second search request for a second item; retrieving a plurality of second reference information based on the second search request; obtaining one or more pieces of second reference information to be used in generating second output information based on a result of the search based on the second search request; generating second input information for the machine learning model based on the information about the second item and the one or more second reference information; generating the second output information by inputting the second input information into the machine learning model; generating third input information for the machine learning model based on the first output information and the second output information; generating third output information by inputting the third input information into the machine learning model; A program for executing a process.

Images