Information processing system and information processing method

Abstract

The present invention provides an information processing system that uses a language model to assist with correction of a document. The information processing system includes an accepting unit and a processing unit. The accepting unit functions to accept document data. The processing unit is configured to execute the following: processing to divide a document included in the document data into a plurality of first blocks; processing to create a prompt including a target sentence containing one of the plurality of first blocks and an instruction sentence containing an instruction for correcting the sentence; processing to acquire at least one second block by transmitting a prompt to a language model over a network, the second block being a proposed correction to the target sentence included in the prompt; processing to update the document data by replacing one of the plurality of first blocks with one of the at least one second block; processing to determine whether correction is necessary for one or both of the first block and the second block; and processing to evaluate the second block.

Description

Information processing system and information processing method

[0001] TECHNICAL FIELD One aspect of the present invention relates to an information processing system and an information processing method, and more particularly, to an information processing system and an information processing method that can utilize processing using a language model.

[0002] One embodiment of the present invention is not limited to the above technical field, and examples of the technical field of one embodiment of the present invention include semiconductor devices, display devices, light-emitting devices, power storage devices, memory devices, electronic devices, lighting devices, input devices (e.g., touch sensors), input / output devices (e.g., touch panels), driving methods thereof, and manufacturing methods thereof.

[0003] When creating a document, not only do you need to check or correct typos and omissions in the text, but you also need to check whether the written content is correct, which requires a lot of effort. In recent years, systems that support document checking and correction have been used. Patent Document 1 discloses a document proofreading device that detects typos and omissions using an error detection model built by machine learning.

[0004] In recent years, the development of language models using neural networks has been actively carried out, and large-scale language models (LLMs) in particular have attracted attention. A large-scale language model is a natural language processing model trained using a large amount of data. A large-scale language model can realize a dialogue model that responds to user instructions, for example. Non-Patent Document 1 discloses GPT-4 (Generative Pre-trained Transformer 4) (registered trademark) as a large-scale language model, and also discloses ChatGPT as a dialogue model.

[0005] The use of large-scale language models has significantly increased the capabilities of natural language processing models. However, as language models become larger, it is difficult to incorporate and operate language models in-house due to the equipment and cost involved. Therefore, one way to use language models is to use external services that provide language models.

[0006] JP 2019-145023 A

[0007] Summary of ChatGPT / GPT-4 Research and Perspective Towards the Future of Large Language Models, Yiheng Liu et al. (Submitted on 4 Apr 2023, [online], Internet <URL: https: / / arxiv.org / abs / 2304.01852>

[0008] Recent improvements in the performance of language models have made it possible to use language models to correct documents. However, correcting documents using a language model requires instructing the language model to make corrections, confirming the corrections, and reflecting them in the document, which is time-consuming. Therefore, while system support is effective, it is necessary to manage corrections made using the language model.

[0009] In view of the above, it is an object of one embodiment of the present invention to provide an information processing system or an information processing method that supports document correction by utilizing a language model.It is an object of one embodiment of the present invention to provide an information processing system or an information processing method that manages document correction work by utilizing a language model.It is an object of one embodiment of the present invention to provide a highly convenient information processing system or an information processing method.It is an object of one embodiment of the present invention to provide a novel information processing system or an information processing method.

[0010] Note that the description of these problems does not preclude the existence of other problems. One embodiment of the present invention does not necessarily have to solve all of these problems. Problems other than these can be extracted from the description in the specification, drawings, and claims.

[0011] One aspect of the present invention is an information processing system including a receiving unit and a processing unit. The receiving unit has a function of receiving document data. The processing unit is configured to execute the following steps: divide a document included in the document data into a plurality of first blocks; create a prompt including a target sentence including one of the plurality of first blocks and an instruction statement including an instruction to correct the sentence; obtain at least one second block that is a correction proposal for the target sentence included in the prompt by transmitting the prompt to a language model via a network; update the document data by replacing one of the plurality of first blocks with at least one second block; determine whether correction is necessary for one or both of the first block and the second block; and evaluate the second block based on the expression, format, or form of the sentence.

[0012] In the above information processing system, the document data is preferably a specification belonging to a patent application.

[0013] The information processing system preferably has an output unit, and the output unit preferably has a function of outputting updated document data.

[0014] In the above information processing system, it is preferable that the determination of whether or not correction is necessary is made using a determination model, the processing unit is configured to execute processing using the determination model, and the determination model is a different type of model from the language model.

[0015] In the above information processing system, it is preferable that the evaluation is performed using an evaluation model, the processing unit is configured to execute processing using the evaluation model, and the evaluation model is a different type of model from the language model.

[0016] The information processing system preferably includes a storage unit, and the storage unit preferably has a function of storing the instruction sentence in the prompt.

[0017] One aspect of the present invention is an information processing method that receives document data, divides a document included in the document data into a plurality of first blocks, determines whether one of the plurality of first blocks needs to be corrected, creates a prompt that includes a target sentence that includes one of the plurality of first blocks and an instruction statement that includes an instruction to correct the sentence, transmits the prompt to a language model via a network to obtain at least one second block that is a suggested correction to the target sentence included in the prompt, evaluates each of the at least one second block based on the expression, format, or form of the sentence, and updates the document data by replacing one of the plurality of first blocks with one of at least one second block.

[0018] In the above information processing method, it is preferable to determine whether or not correction is required for each of at least one second block.

[0019] In the above information processing method, the document data is preferably a specification belonging to a patent application.

[0020] In the above information processing method, it is preferable that the determination of whether or not correction is necessary is performed using a determination model, and that the determination model is a model of a different type from the language model.

[0021] In the above information processing method, the evaluation is performed using an evaluation model, and the evaluation model is preferably a model of a different type from the language model.

[0022] One aspect of the present invention is a program having a function of causing a processor to execute any of the above information processing methods.

[0023] According to one aspect of the present invention, an information processing system or an information processing method can be provided that supports document correction using a language model. According to one aspect of the present invention, an information processing system or an information processing method can be provided that manages document correction work using a language model. According to one aspect of the present invention, a highly convenient information processing system or an information processing method can be provided. According to one aspect of the present invention, a novel information processing system or an information processing method can be provided.

[0024] Note that the description of these effects does not preclude the existence of other effects. One embodiment of the present invention does not necessarily have all of these effects. Effects other than these can be extracted from the description in the specification, drawings, and claims.

[0025] FIG. 1 is a diagram illustrating an example of an information processing system. FIG. 2 is a diagram illustrating an example of an information processing system. FIG. 3 is a block diagram illustrating an example of an information processing system. FIG. 4 is a diagram illustrating an example of an information processing method. FIGS. 5A to 5C are diagrams illustrating an example of an information processing method. FIGS. 6A and 6B are diagrams illustrating an example of an information processing method. FIGS. 7A and 7B are diagrams illustrating an example of an information processing method. FIGS. 8A and 8B are diagrams illustrating an example of an information processing method. FIG. 9 is a diagram illustrating an example of an information processing system. FIG. 10 is a block diagram illustrating an example of an information processing system. FIG. 11 is a diagram illustrating an example of an information processing system. FIG. 12 is a diagram illustrating an example of an information processing system.

[0026] The embodiments will be described in detail with reference to the drawings. However, the present invention is not limited to the following description, and it will be readily understood by those skilled in the art that various changes can be made in form and detail without departing from the spirit and scope of the present invention. Therefore, the present invention should not be interpreted as being limited to the description of the embodiments shown below.

[0027] In the configuration of the invention described below, the same parts or parts having similar functions are denoted by the same reference numerals in different drawings, and repeated explanations thereof will be omitted. Furthermore, when referring to similar functions, the same hatching pattern may be used and no particular reference numeral may be assigned.

[0028] Furthermore, for ease of understanding, the position, size, range, etc. of each component shown in the drawings may not represent the actual position, size, range, etc. Therefore, the disclosed invention is not necessarily limited to the position, size, range, etc. disclosed in the drawings.

[0029] In this specification, the ordinal numbers "first" and "second" are used for convenience and do not limit the number of components or the order of the components (for example, the order of processes or the order of stacking). Furthermore, the ordinal numbers assigned to components in one part of this specification may not match the ordinal numbers assigned to the same components in other parts of this specification or in the claims.

[0030] In this specification, when the same symbol is used for multiple elements, and particularly when it is necessary to distinguish between them, an identification symbol such as “_1”, “[n]”, or “[m, n]” may be added to the symbol.

[0031] Embodiment In this embodiment, an information processing system according to one embodiment of the present invention will be described with reference to FIGS.

[0032] An information processing system according to an aspect of the present invention includes a receiving unit and a processing unit. The receiving unit has a function of receiving document data.

[0033] The documents included in the document data accepted by the information processing system of one embodiment of the present invention are not particularly limited, and the information processing system of one embodiment of the present invention can support the correction of various types of documents. Examples of such documents include patent application documents, books, magazines, newspapers, contracts, papers (including academic papers, theses, doctoral dissertations, essays, journal articles, etc.), court decisions, terms and conditions, product manuals, novels, publications, white papers, technical documents, and business documents. More specifically, patent application documents include one or more of the specification, claims, and abstract of a patent application.

[0034] Checking and correcting documents is extremely laborious. Furthermore, because preferred expressions differ depending on the type of document, users must be aware of the type of document when making corrections. Furthermore, the time required for corrections can be long depending on the content of the document and the user's level of experience or proficiency.

[0035] Therefore, the processing unit is configured to execute a process of dividing a document included in the document data into a plurality of first blocks and a process of determining whether each of the first blocks needs to be corrected. By employing such a configuration, a user can easily find character strings that need to be corrected.

[0036] The information processing system of this embodiment is capable of transmitting and receiving data via a network with an information processing device capable of performing processing using a language model. The processing unit is configured to execute a process of creating a correction prompt including the first block as a target sentence and a process of acquiring at least one second block by transmitting the correction prompt to the language model via the network. This allows the user to easily obtain a correction suggestion for the first block.

[0037] The processing unit is configured to execute a process for evaluating the second block. Specifically, it is preferable to evaluate the second block based on the expression, format, or form of the sentence. For example, appropriate sentence endings differ depending on the classification of the document. This configuration makes it possible to present the user with preferred suggestions for modifying the expression.

[0038] The processing unit is configured to execute processing for updating the document data by replacing the first block with the second block. With this configuration, it is possible to correct a document included in the document data.

[0039] With the above-described configuration, it is possible to realize an information processing system that uses a language model to assist in document correction.

[0040] The information processing system described above can also be regarded as an information processing system capable of executing processing using an information processing method according to an embodiment of the present invention. Specifically, one embodiment of the present invention is an information processing method that receives document data, divides a document included in the document data into a plurality of first blocks, determines whether at least one of the plurality of first blocks needs to be corrected, creates a prompt including at least one of the plurality of first blocks as a target sentence, transmits the prompt to a language model via a network to obtain at least one second block, evaluates each of the at least one second block, and updates the document data by replacing at least one of the plurality of first blocks with one of the at least one second block.

[0041] <Configuration Example 1 of Information Processing System> Fig. 1 shows an example of an information processing system according to this embodiment. The information processing system according to this embodiment includes an information processing device 10, an information terminal 20, and an information processing device 40. In the information processing system shown in Fig. 1, the information processing device 10 is connected to the information processing device 40 via a network 30. The information processing device 10 is also connected to a plurality of information terminals 20 via a network 31. Fig. 1 shows information terminals 20a, 20b, 20c, and 20d as examples of the information terminals 20.

[0042] A user of the information processing system can access the information processing device 10 from the information terminals 20a to 20d, etc. The user can then receive a service using the information processing method according to one aspect of the present invention.

[0043] A user can use the information processing device 10, for example, by using dedicated software or an application installed on the information terminal 20. Alternatively, a user can use the information processing device 10, for example, from a web browser on the information terminal 20.

[0044] An organization such as a company that provides a service using an information processing method according to one embodiment of the present invention (hereinafter also referred to as a service provider) can provide the service using the information processing device 10. The service provider does not need to prepare the information processing device 40 on its own, but can use the information processing device 40 via the network 30.

[0045] The information processing device 10 is a device that can execute processing using an information processing method according to one aspect of the present invention. In FIG. 1, a server computer is illustrated as an example.

[0046] The information processing device 10 can perform information processing such as calculations using data input from the information terminal 20a via the network 31. The information processing device 10 can transmit the results of the information processing to the information terminal 20a via the network 31. This can reduce the calculation load on the information terminal 20a. Here, the information terminal 20a has been described as an example, but the same can be said for the information terminals 20b to 20d.

[0047] The information processing device 10 can input data to the information processing device 40 via the network 30. The information processing device 40 can perform information processing such as calculations using the input data. The information processing device 40 can transmit the results of the information processing to the information processing device 10 via the network 30. This can reduce the calculation load on the information processing device 10. The information processing device 10 can then transmit the data received from the information processing device 40 or data created based on the data to the information terminal 20a via the network 31.

[0048] The information terminals 20a to 20d are each an information terminal device such as a computer used by a user, and may also be referred to as a client computer. FIG. 1 illustrates, as an example, the information terminal 20a, which is a desktop computer; the information terminal 20b, which is a notebook computer; the information terminal 20c, which is a smartphone; and the information terminal 20d, which is a tablet computer. The information terminal 20d can also be used as a notebook computer by connecting it to a housing 21 having a keyboard. The number of information terminals connected to the information processing device 10 is not particularly limited. While FIG. 1 shows four information terminals, the number of information terminals may be one, two, three, five, or more. Examples of such information terminals include desktop information terminals, notebook information terminals, tablet information terminals, and mobile information terminals such as smartphones.

[0049] The information processing device 10 is a large computer such as a server computer or a supercomputer. The information processing device 10 is a computer with higher processing power than the information terminal 20. The information processing device 10 can function as a parallel computer. By using the information processing device 10 as a parallel computer, it is possible to perform large-scale calculations required for learning and inference of artificial intelligence (AI), for example. The information processing device 10 may also be capable of performing processing using a natural language processing model using AI.

[0050] The information processing device 40 is a large computer such as a server computer or a supercomputer. The information processing device 40 is a computer with higher processing power than the information terminal 20. The information processing device 40 is also a computer with higher processing power than the information processing device 10. It is preferable that the information processing device 40 has a function as a parallel computer. By using the information processing device 40 as a parallel computer, it is possible to perform large-scale calculations required for AI learning and inference, for example. Note that when both the information processing device 10 and the information processing device 40 have a function as a parallel computer, the information processing device 40 has a higher processing power and can perform large-scale calculations than the information processing device 10.

[0051] The information processing device 40 can perform processing using a natural language processing model that uses AI. Examples of the natural language processing model that uses AI include BERT (Bidirectional Encoder Representations from Transformers) and T5 (Text-to-Text Transfer Transformer). The information processing device 40 can also perform processing using a model that utilizes a large-scale language model. Examples of large-scale language models include GPT-3, GPT-3.5, GPT-4 (registered trademark), LaMDA (Language Model for Dialogue Applications), PaLM (Pathways Language Model), and PaLM2, and it is preferable to use GPT-4 (registered trademark).

[0052] The information processing device 40 can execute processing using a general-purpose AI model that can perform various tasks. When the information processing device 10 and the information processing device 40 can each execute processing using a model that uses a large-scale language model, the information processing device 40 can execute processing using a larger-scale AI model than the information processing device 10.

[0053] Network 31 is a computer network smaller than network 30. Typically, network 30 is a global network, and network 31 is a local network. It is preferable to use the Internet, which is the foundation of the World Wide Web (WWW), as network 30. It is preferable to use an intranet or an extranet as network 31. In this embodiment, an example will be mainly described in which the Internet is used as network 30 and an intranet is used as network 31.

[0054] Other computer networks that can be used as the network 31 include a PAN (Personal Area Network), a LAN (Local Area Network), a CAN (Campus Area Network), a MAN (Metropolitan Area Network), a WAN (Wide Area Network), and a GAN (Global Area Network).

[0055] Furthermore, when wireless communication is performed, communication standards such as the fourth generation mobile communication system (4G), fifth generation mobile communication system (5G), and sixth generation mobile communication system (6G), or specifications standardized by the IEEE such as Wi-Fi (registered trademark) and Bluetooth (registered trademark), can be used as communication protocols or communication technologies.

[0056] The information processing system shown in FIG. 1 is suitable for a case where a user who uses the information terminal 20 and a service provider belong to the same organization such as a company.

[0057] For example, it is preferable that data is transmitted and received between the information terminal 20 and the information processing device 10 using a network 31 established within an organization such as a company. It is also preferable that data is transmitted and received between the information processing device 10 and the information processing device 40 using the Internet. This allows applications, tools, services, etc. provided outside the organization to be used more safely than when the information terminal 20 is directly connected to the information processing device 40 via the Internet.

[0058] The flow of data in the information processing system shown in Fig. 1 will be described with reference to Fig. 2. In Fig. 2, the order of data transmission and reception is indicated by circled numbers.

[0059] First, the user inputs desired data DA1 into the information processing device 10 using the information terminal 20. The data DA1 is transmitted from the information terminal 20 to the information processing device 10 via the network 31.

[0060] The data DA1 is document data corresponding to the document to be corrected. The format of the data DA1 is not particularly important, but it is preferably text data. Details of the data DA1 will be explained later in the section <Information Processing Method>.

[0061] When the data DA1 includes data other than text data, it is preferable to convert the data DA1 into text data in the information processing device 10. When converting various data into text data, it is preferable to use AI technology. The following describes an example in which the data DA1 is text data. Note that when the data DA1 includes data other than text data, the data DA1 can be read as text data generated using the data DA1.

[0062] The information processing device 10 preferably has a function of generating text data using image data. For example, the information processing device 10 can generate text data using image data by performing processing using at least one AI technology, such as image recognition, handwritten character recognition, or caption generation. The information processing device 10 may also have an optical character recognition (OCR) function.

[0063] Next, the information processing device 10 creates data DA2 based on the data DA1. The data DA2 corresponds to text data equivalent to a prompt that the user wants to input to the language model. The prompt corresponds to an input sentence that causes the language model to perform a desired operation. When a prompt is given, the language model generates a response content based on the prompt. Details of the data DA2 will be explained later in the section <Information Processing Method>.

[0064] The information processing device 10 can create data DA2 by adding second text data to first text data. Here, the first text data is part of the text data included in data DA1, and the second text data is text data corresponding to an instruction sentence in a prompt to be input to a language model. Here, the instruction sentence is part of the prompt to be input to a language model, and can be rephrased as a question sentence, a command sentence, or the like.

[0065] The second text data may be a system prompt. In this specification, a system prompt refers to an instruction sentence in a prompt that is preset by a developer or a service provider. Presetting a system prompt makes it easier to obtain a desired answer from a language model.

[0066] An example of the second text data is an instruction sentence including a command to correct a sentence, such as "Please correct target sentence A." Note that the text data corresponding to target sentence A is the first text data.

[0067] The data DA2 obtained through such processing is transmitted from the information processing device 10 to the information processing device 40 via the network 30.

[0068] Next, data DB1 is transmitted from information processing device 40 to information processing device 10 via network 30. That is, information processing device 10 acquires data DB1 from information processing device 40 via network 30. Data DB1 is text data equivalent to a response sentence corresponding to data DA2 input to information processing device 40. Details of data DB1 will be explained later in <Information Processing Method>.

[0069] Next, the information processing device 10 creates a data DB2 based on the data DB1. The data DB2 is the data DA1 updated based on the data DB1. Details of the data DB2 will be described later in the section <Information Processing Method>.

[0070] Next, the information processing device 10 outputs the data DB2 to the information terminal 20 via the network 31. As a result, the user can obtain the data DB2 created by updating the data DA1.

[0071] For example, data DB2 may be converted into data other than text data and then output to information terminal 20. For example, if data DA1 is data other than text data, data DB2 is preferably data in the same format as data DA1. Information processing device 10 may have a function of generating image data using text data, for example.

[0072] Fig. 3 is a block diagram of the information processing system shown in Fig. 1. In Fig. 3, an example of the configuration of the information processing device 10 will be mainly described.

[0073] The information processing device 10 includes a receiving unit 110 , a storage unit 120 , a processing unit 130 , an output unit 140 , and a transmission path 150 .

[0074] In the drawings accompanying this specification, the components are classified by function and shown as independent blocks in the block diagrams. However, in reality, it is difficult to completely separate the components by function, and one component may be involved in multiple functions. For example, part of the processing unit 130 may function as the reception unit 110. Also, one function may be involved in multiple components. For example, the processing performed by the processing unit 130 may be executed by different servers depending on the processing.

[0075] [Reception Unit 110] The reception unit 110 can receive data from outside the information processing device 10. For example, the reception unit 110 receives data from the information terminal 20 via the network 31. The data received from the information terminal 20 corresponds to the data DA1 described above. Furthermore, for example, the reception unit 110 receives data from the information processing device 40 via the network 30. The data received from the information processing device 40 corresponds to the data DB1 described above.

[0076] As described above, it is preferable that the data accepted by the accepting unit 110 is text data. Furthermore, if the data accepted by the accepting unit 110 is data other than text data, it is preferable that the processing unit 130 generates text data using the data accepted by the accepting unit 110.

[0077] The data supplied to the reception unit 110 is supplied to one or both of the storage unit 120 and the processing unit 130 via the transmission path 150 .

[0078] [Storage Unit 120] The storage unit 120 has a function of storing a program executed by the processing unit 130. The storage unit 120 may also have a function of storing data created by the processing unit 130 (e.g., calculation results, analysis results, inference results), data input to the reception unit 110, etc.

[0079] The storage unit 120 preferably has a function of storing instruction sentences in the prompts. Details of the instruction sentences in the prompts will be explained later in the section <Information Processing Method>.

[0080] The storage unit 120 may have a database. Furthermore, the information processing device 10 may have a database separate from the storage unit 120. The information processing device 10 may have a function to retrieve data from a database that exists outside the storage unit 120, outside the information processing device 10, or outside the information processing system. Furthermore, the information processing device 10 may have a function to retrieve data from both its own database and an external database.

[0081] A file server may be used instead of the database. For example, when using files stored in a file server, it is preferable that the database has paths to files stored in the file server.

[0082] The storage unit 120 includes at least one of a volatile memory and a non-volatile memory. Examples of the volatile memory include a dynamic random access memory (DRAM) and a static random access memory (SRAM). Examples of the non-volatile memory include a resistive random access memory (ReRAM), a phase change random access memory (PRAM), a ferroelectric random access memory (FeRAM), a magnetoresistive random access memory (MRAM), and a flash memory. The storage unit 120 may also include at least one of NOSRAM (registered trademark) and DOSRAM (registered trademark). The storage unit 120 may also include a recording media drive. Examples of recording media drives include a hard disk drive (HDD) and a solid state drive (SSD).

[0083] NOSRAM is an abbreviation for "Nonvolatile Oxide Semiconductor Random Access Memory (RAM)." NOSRAM refers to a memory in which memory cells are two-transistor (2T) or three-transistor (3T) gain cells and transistors (also called OS transistors) that use metal oxide in their channel formation regions. OS transistors have an extremely small leakage current, i.e., a current that flows between the source and drain in an off state. NOSRAM can be used as a nonvolatile memory by retaining a charge corresponding to data in the memory cell using its extremely small leakage current characteristic. In particular, NOSRAM can read stored data without destroying it (nondestructive readout), making it suitable for arithmetic processing in which only data read operations are repeated a large number of times. NOSRAM can increase its data capacity by stacking layers, and therefore can be used as a large-scale cache memory, main memory, or storage memory to improve the performance of semiconductor devices.

[0084] DOSRAM is an abbreviation for "Dynamic Oxide Semiconductor RAM" and refers to a RAM having 1T (transistor) 1C (capacitor) type memory cells. DOSRAM is a DRAM formed using OS transistors, and is a memory that temporarily stores information sent from an external device. DOSRAM is a memory that takes advantage of the low off-state current of OS transistors.

[0085] In this specification and the like, a metal oxide refers to an oxide of a metal in a broad sense. Metal oxides are classified into oxide insulators, oxide conductors (including transparent oxide conductors), oxide semiconductors (also referred to as oxide semiconductors or simply as OSs), and the like. For example, when a metal oxide is used for a semiconductor layer of a transistor, the metal oxide may be referred to as an oxide semiconductor.

[0086] The metal oxide included in the channel formation region preferably contains indium (In). When the metal oxide included in the channel formation region contains indium, the carrier mobility (electron mobility) of the OS transistor is increased. Furthermore, the metal oxide included in the channel formation region is preferably an oxide semiconductor containing element M. The element M is preferably at least one of aluminum (Al), gallium (Ga), and tin (Sn). Other elements applicable to element M include boron (B), silicon (Si), titanium (Ti), iron (Fe), nickel (Ni), germanium (Ge), yttrium (Y), zirconium (Zr), molybdenum (Mo), lanthanum (La), cerium (Ce), neodymium (Nd), hafnium (Hf), tantalum (Ta), and tungsten (W). However, a combination of two or more of the above elements may be used as element M. The element M is, for example, an element having a high bond energy with oxygen. For example, it is an element having a higher bond energy with oxygen than indium. The metal oxide contained in the channel formation region is preferably a metal oxide containing zinc (Zn), since zinc-containing metal oxides may be easily crystallized.

[0087] The metal oxide contained in the channel formation region is not limited to a metal oxide containing indium, but may be, for example, a metal oxide containing zinc but not indium, such as zinc tin oxide or gallium tin oxide, a metal oxide containing gallium, or a metal oxide containing tin.

[0088] [Processing Unit 130] The processing unit 130 has a function of performing processes such as calculation, analysis, and inference using data supplied from one or both of the receiving unit 110 and the storage unit 120. The processing unit 130 can supply the created data (e.g., calculation results, analysis results, and inference results) to one or both of the storage unit 120 and the output unit 140.

[0089] The processing unit 130 has a function of acquiring data from the storage unit 120. The processing unit 130 may also have a function of recording or registering data in the storage unit 120.

[0090] The processing unit 130 has a function of dividing a document included in the document data into multiple blocks. For example, the processing unit 130 has a function of dividing a document included in the document data into multiple first blocks. One block includes at least one sentence. For example, a document can be divided into blocks, each consisting of one paragraph. The document may also be divided using information such as headings included in the document. Alternatively, blocks may be divided by sentence, i.e., each block may consist of one sentence. The document data is data received by the receiving unit 110 and corresponds to data DA1.

[0091] The processing unit 130 has a function of creating a prompt that includes at least one of the multiple blocks as a target sentence. For example, the processing unit 130 has a function of creating a prompt that includes at least one of the multiple first blocks as a target sentence. The created prompt corresponds to data DA2.

[0092] The processing unit 130 has a function of acquiring at least one block by transmitting the prompt to the language model via the transmission path 150, the output unit 140, and the network 30. For example, the processing unit 130 has a function of acquiring a response sentence corresponding to the prompt by transmitting the prompt to the language model via the transmission path 150, the output unit 140, and the network 30, and extracting at least one second block from the response sentence. The acquired response sentence corresponds to data DB1. Note that, because the extracted second block is included in the acquired response sentence, the second block can also be said to correspond to data DB1. Hereinafter, the process of acquiring a response sentence corresponding to a prompt by transmitting it to the language model and extracting at least one block from the response sentence may be referred to as the process of acquiring at least one block by transmitting a prompt to the language model.

[0093] The processing unit 130 has a function of updating the document data by replacing at least one of the plurality of first blocks with one of at least one of the second blocks. Note that at least one of the plurality of first blocks is a block included in the target sentence of the prompt. Therefore, for example, the processing unit 130 has a function of updating the document data by replacing the block that is the target sentence of the prompt with one of the at least one of the acquired blocks. The updated document data corresponds to the data DB2.

[0094] The processing unit 130 has a function of determining whether a block needs to be corrected. For example, the processing unit 130 has a function of determining whether one or both of the first block and the second block needs to be corrected. The determination of whether the first block needs to be corrected may be performed before creating a prompt to be sent to the language model. Furthermore, for example, the determination of whether the second block needs to be corrected may be performed after acquiring the second block from the language model. A method for determining whether a block needs to be corrected will be described later in the section <Information Processing Method>.

[0095] The determination of whether or not correction is necessary is preferably performed using a determination model. That is, the processing unit 130 preferably has a function of determining whether or not correction is necessary for a block using the determination model. For example, the processing unit 130 preferably has a function of determining whether or not correction is necessary for one or both of the first block and the second block using the determination model.

[0096] The processing unit 130 has a function of evaluating blocks. For example, the processing unit 130 has a function of evaluating a second block. Note that the target of evaluation is not limited to the second block, and may be the first block. The evaluation method will be described later in the section <Information Processing Method>.

[0097] The evaluation is preferably performed using an evaluation model. That is, the processing unit 130 preferably has a function of evaluating the block using the evaluation model. For example, the processing unit 130 preferably has a function of evaluating the second block using the evaluation model.

[0098] The processing unit 130 may include, for example, an arithmetic circuit, a central processing unit (CPU), and a graphics processing unit (GPU).

[0099] The processing unit 130 may have a microprocessor such as a DSP (Digital Signal Processor). The microprocessor may be implemented by a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array) or an FPAA (Field Programmable Analog Array). The processing unit 130 may also have a quantum processor. The processing unit 130 can perform various data processing and program control by interpreting and executing instructions from various programs using the processor. Programs that can be executed by the processor are stored in at least one of the memory area of ​​the processor and the storage unit 120.

[0100] The processing unit 130 may include a main memory. The main memory may include at least one of a volatile memory such as a random access memory (RAM) and a non-volatile memory such as a read-only memory (ROM). The main memory may also include at least one of the above-mentioned NOSRAM and DOSRAM.

[0101] The RAM may be, for example, a DRAM or an SRAM, and a virtual memory space is allocated and used as a working space for the processing unit 130. The operating system, application programs, program modules, program data, lookup tables, and the like stored in the storage unit 120 are loaded into the RAM for execution. The data, programs, and program modules loaded into the RAM are each directly accessed and operated by the processing unit 130.

[0102] The ROM can store a BIOS (Basic Input / Output System), firmware, etc., which do not require rewriting. Examples of ROM include mask ROM, OTPROM (One Time Programmable Read Only Memory), and EPROM (Erasable Programmable Read Only Memory). Examples of EPROMs include UV-EPROMs (Ultra-Violet Erasable Programmable Read Only Memories), which allow stored data to be erased by exposure to ultraviolet light, EEPROMs (Electrically Erasable Programmable Read Only Memories), and flash memories.

[0103] The processing section 130 can include one or both of an OS transistor and a transistor having silicon in a channel formation region (a Si transistor).

[0104] The processing unit 130 preferably includes an OS transistor. Because an OS transistor has an extremely small off-state current, using the OS transistor as a switch for retaining charge (data) flowing into a capacitor functioning as a memory element can ensure a long data retention period. By using this characteristic in at least one of the register and cache memory of the processing unit, the processing unit can be operated only when necessary, and can be turned off in other cases by saving information from the previous processing in the memory element. In other words, normally-off computing is possible, and the power consumption of the information processing system can be reduced.

[0105] It is preferable that the information processing device 10 uses AI for at least some of its processing.

[0106] In particular, it is preferable that the information processing device 10 uses an artificial neural network (ANN, hereinafter also simply referred to as a neural network). A neural network is realized by a circuit (hardware) or a program (software).

[0107] In this specification, a neural network refers to a general model that mimics the neural circuit network of a living organism, determines the connection strength between neurons through learning, and has problem-solving capabilities. A neural network has an input layer, an intermediate layer (hidden layer), and an output layer.

[0108] In this specification and the like, when discussing neural networks, determining the connection strengths (also called weighting coefficients) between neurons from existing information may be referred to as "learning."

[0109] In this specification and the like, the act of constructing a neural network using connection strengths obtained by learning and deriving a new conclusion from it may be referred to as "inference."

[0110] [Output Unit 140] The output unit 140 outputs data based on the processing result of the processing unit 130. The output unit 140 can supply at least one of the calculation result, analysis result, and inference result of the processing unit 130 to an external device of the information processing device 10.

[0111] The output unit 140 can output data to the outside of the information processing device 10. For example, the output unit 140 outputs data to the information terminal 20 via the network 31. The output unit 140 can output data created by the processing unit 130 based on data received from the information processing device 40 to the information terminal 20. Alternatively, the output unit 140 can output the data itself received from the information processing device 40 to the information terminal 20. The data output to the information terminal 20 corresponds to the data DB 2 described above.

[0112] The data can be presented, for example, by displaying it on the display screen of the information terminal 20, or by outputting a file in a format such as txt, docx, xml, pdf, or csv.

[0113] Furthermore, the output unit 140 outputs data to the information processing device 40 via the network 30. The output unit 140 outputs data created by the processing unit 130 based on data received from the information terminal 20 to the information processing device 40. The data output to the information processing device 40 corresponds to the data DA2 described above.

[0114] [Transmission Path 150] The transmission path 150 has a function of transmitting data. Data can be transmitted and received between the reception unit 110, the storage unit 120, the processing unit 130, and the output unit 140 via the transmission path 150.

[0115] <Information Processing Method> The information processing method of this embodiment includes steps S11 to S19 shown in FIG. 4. FIGS. 5A to 8B are diagrams illustrating the information processing method. FIGS. 6A to 8B can be considered examples of a graphical user interface (GUI) related to the information processing system of this embodiment. Forms, icons, tables, etc. in the drawings related to the GUI illustrated in this embodiment, such as FIG. 6A, are examples and are not particularly limited. The GUI can be configured as a web page that a user accesses via a network. Alternatively, the GUI can be configured as a screen of a program application executed on a terminal used by the user. FIGS. 6A to 8B show an example of instructing a language model to proofread a block (sentence or document).

[0116] [Step S11] In step S11, the receiving unit 110 receives document data. Specifically, the receiving unit 110 receives the document data via the information terminal 20. The document data corresponds to the above-mentioned data DA1. In the information processing method of this embodiment, a case will be described in which the receiving unit 110 receives document data including the document 301.

[0117] For example, the user inputs document data into the information processing device 10 using the information terminal 20. Alternatively, for example, the user may specify document data stored in the storage unit 120. Then, the document 301 included in the document data is output from the information processing device 10 to the display unit of the information terminal 20.

[0118] [Step S12] In step S12, the processing unit 130 divides the document 301 included in the document data into a plurality of blocks 302. The blocks 302 correspond to the first blocks described above.

[0119] For example, as shown in Fig. 5A, a document 301 is divided into a plurality of blocks 302 by the processing unit 130. Fig. 6A shows an example in which the document 301 is divided into paragraphs, and the paragraph numbers of the target blocks are displayed on the display unit of the information terminal 20. Specifically, a table including paragraphs 321 and blocks 322 is displayed on the display unit. Note that the items in the table are not limited to these.

[0120] In the information processing method of this embodiment, a case will be described in which a document 301 is divided into n blocks 302 (blocks 302[1] to 302[n]) (n is an integer of 2 or more).

[0121] In addition, if document 301 is the smallest unit to be divided in step S12 (for example, if document 301 consists of one paragraph when dividing by paragraphs, or if document 301 consists of one sentence when dividing by sentences), the processing of step S12 can be omitted.

[0122] The user selects one of the multiple blocks 302 output on the display unit of the information terminal 20. For example, as shown in Fig. 6A, the user selects a block that requires consideration for revision or the paragraph number of that block with the mouse pointer 324. Note that in Fig. 6A, as an example, the selection of paragraph 0001 is indicated by hatching paragraph 0001. That is, in Fig. 6A, block 302[1] is selected.

[0123] If the process of step S12 is omitted, the document 301 can be automatically selected.

[0124] [Step S13] In step S13, the processing unit 130 determines whether or not the selected block needs to be corrected and labels it. In Fig. 6A, the block 302[1] corresponds to the selected block.

[0125] For example, a character string determined to require correction is detected by determining whether the selected block needs correction. Note that a character string determined to require correction can be said to be a character string that contains a typographical error or is likely to contain a typographical error. In this way, by determining whether correction is required, the user can easily find a character string that contains a typographical error or is likely to contain a typographical error.

[0126] Examples of errors that are subject to the determination of whether or not they require correction include incorrect conversions, typos, mistakes between particles and auxiliary verbs, an excess or deficiency of one or both of a period and a comma, a missing subject, an error in a passive or active expression, etc. If the document is a patent application, a paper, a product manual, or a technical document, examples of errors that are subject to the determination of whether or not they require correction include an incomplete list of materials, an incomplete description of measurement conditions, an incomplete description of numerical limitations or numerical ranges, a restrictive description, etc.

[0127] Furthermore, labeling is performed on character strings that are determined to require correction. For example, labels are assigned based on the type of typo. For example, if a character string that is determined to require correction is detected as a mistranslation, labeling can be performed to assign a label corresponding to the mistranslation to the character string. In this way, by performing labeling, the user can easily confirm which type of mistranslation the detected character string corresponds to.

[0128] The determination of whether correction is necessary and the labeling are preferably performed using a determination model 210. The determination model 210 preferably has a function of detecting character strings that require correction. Furthermore, the determination model 210 preferably has a function of assigning a label to the detected character strings.

[0129] Furthermore, it is preferable that a model be prepared for each type of clerical error as the determination model 210. By preparing a model for each type of clerical error, the accuracy of determining whether correction is necessary can be improved. Furthermore, the amount of calculation can be reduced. For example, FIG. 5B shows a case where three models (model 210a, model 210b, and model 210c) are prepared as the determination model 210. Note that the number of models prepared as the determination model 210 is not limited to three, and may be one, two, four or more.

[0130] The determination model 210 is preferably a model optimized by machine learning or a rule-based model. In this case, the determination model 210 can be said to be a different type of model from a language model. By using an optimized model or a rule-based model, high accuracy can be obtained with a smaller amount of calculation (fewer parameters) than a language model. The determination model 210 may also be a model that combines machine learning and rule-based models. The determination model 210 may also be configured by combining a language model and a prompt.

[0131] For example, the determination of whether or not a selected block needs to be corrected and labeling are performed using the determination model 210. In Fig. 5B, the determination of whether or not a block 302[1] needs to be corrected and labeling are performed using the determination model 210. Fig. 6B shows a case where a first character string and a second character string are detected by determining whether or not a block 302[1] needs to be corrected. The first character string and the second character string are character strings that are determined to need correction.

[0132] It is preferable that character strings determined to require correction be highlighted. By highlighting the character strings, the user can easily visually recognize the character strings. In FIG. 6B , the first character string and the second character string are highlighted by hatching. Note that the highlighting method is not limited to hatching, and examples of the highlighting method include changing the background color, changing the character color, increasing the character size, changing the font, making the character bold, and underlining the character strings determined to require correction.

[0133] In the information processing method of this embodiment, the first character string and the second character string are assumed to have different types of errors. For example, the first character string is detected as a first error, and the second character string is detected as a second error. Furthermore, as shown in FIG. 5B , if multiple models are prepared as the determination model 210, for example, the first character string is detected as a first error by model 210a, and the second character string is detected as a second error by model 210b. In this case, by performing labeling using the determination model 210, different labels are assigned to the first character string and the second character string. In FIG. 6B , the first character string and the second character string are highlighted with different hatching. Specifically, the first character string is indicated by hatching 311a, and the second character string is indicated by hatching 311b. 6B, text 323a explaining the first type of error is displayed with hatching applied to the first character string outside the table including paragraph 321 and block 322. Similarly, text 323b explaining the second type of error is displayed with hatching applied to the second character string.

[0134] The user determines whether or not there is a character string that needs to be corrected among the detected character strings (specifically, the highlighted character strings) by determining whether or not there is a character string that needs to be corrected.

[0135] If the user determines that a character string needs to be corrected, the user selects at least one of the character strings detected by the correction determination output on the display unit of the information terminal 20. For example, as shown in FIG. 6B , the user uses the mouse pointer 324 to select a character string determined to need correction from among the character strings detected by the correction determination. When a first character string is selected as the character string determined to need correction, it is preferable that the hatching applied to the first character string be changed. In FIG. 7A , as an example, the hatching applied to the first character string is changed from hatching 311a to hatching 312. Also, in FIG. 7A , text 325 explaining the selection is displayed outside the table including the paragraph 321 and the block 322, along with the hatching applied to the first character string. After at least one character string is selected from the character strings detected by the correction determination, the process proceeds to step S14.

[0136] If the user determines that no correction is required for any of the detected character strings by determining whether or not correction is required, the user selects a block different from the selected block and proceeds to step S13 described above, or ends the update of the document 301.

[0137] If the result of determining whether the selected block needs to be corrected does not indicate that a character string determined to need to be corrected is found, the user selects a block different from the selected block and proceeds to step S13 described above, or ends the update of the document 301.

[0138] [Step S14] In step S14, the processing unit 130 creates a prompt 303 corresponding to the selected character string. The prompt 303 is a prompt for correcting a block containing the selected character string in accordance with a determined error in the selected character string, and can be called a correction prompt. The prompt 303 corresponds to the above-mentioned data DA2.

[0139] The processing unit 130 preferably has a function of presenting an instruction corresponding to the label attached to the selected character string. The instruction includes an instruction to correct the sentence. For example, if the label indicates an excess or deficiency of one or both of a period and a comma, an instruction such as "Please create a revised sentence by reviewing the use of periods and commas in the target sentence A" is presented. Note that the target sentence A is a block including the selected character string. The processing unit 130 can create a prompt 303 using the block including the selected character string and the presented instruction.

[0140] Furthermore, the instruction preferably includes a description that specifies the number of blocks to be output in step S15. In other words, the instruction preferably includes a description of the number of blocks to be generated. For example, if three blocks are to be generated, the instruction may include a description such as "Please create three correction sentences." Therefore, by including the number of blocks to be generated in the instruction, the language model can be made to generate a response sentence that includes multiple blocks.

[0141] It is also possible to obtain multiple response sentences from the language model. The language model probabilistically calculates subsequent phrases from the position of a phrase. At this time, by adding some fluctuation, the language model can generate multiple response sentences. At this time, multiple blocks may be obtained by extracting blocks from each of the multiple response sentences.

[0142] [Step S15] In step S15, the processing unit 130 acquires at least one block by sending a prompt to the language model via the network 30. For example, as shown in FIG. 5C , the processing unit 130 acquires a response sentence to the prompt 303 by sending a prompt 303 to the language model 250, and extracts at least one block 304 from the response sentence. The response sentence or block 304 corresponds to the above-mentioned data DB1, and the block 304 corresponds to the above-mentioned second block. The block 304 can also be considered a revision proposal for the target sentence A included in the prompt 303. In other words, the block 304 can also be considered a revision proposal for the block including the selected character string.

[0143] The language model can generate a specified number of blocks based on the generation number described in prompt 303. In other words, the specified number of blocks can be obtained by performing processing using the language model.

[0144] In Fig. 7B, when prompt 303 contains a description that the number of blocks to be generated is three, the three generated blocks are displayed on the display unit of information terminal 20. In the information processing method of this embodiment, a case where three blocks (block 304[1], block 304[2], and block 304[3]) are generated will be described. In Fig. 7B, a table including serial numbers 326 and generated blocks 327 is displayed on the display unit of information terminal 20. Note that the items in the table are not limited to these.

[0145] The number of blocks 304 to be acquired (the number of generated blocks 304) may be one, but preferably two or more. By acquiring multiple blocks 304, the number of revision suggestions can be increased. Therefore, revision suggestions can be selected according to the user's preferences or the classification of the document 301.

[0146] [Step S16] In step S16, the processing unit 130 determines whether or not the acquired block needs to be corrected. The acquired block is a block generated by a language model. For a method of determining whether or not a correction is needed, see the description of [Step S13]. As described above in step S13, the processing unit 130 can also label the acquired block.

[0147] FIG. 7B illustrates a case in which a second character string is detected by determining whether block 304[1] needs correction. When a prompt to correct the first character string is sent to the language model for block 302[1] containing a first character string and a second character string, the acquired block 304[1] may contain the second character string. Note that the language model does not necessarily output a block in which only the first character string is corrected. For example, as in block 304[2] shown in FIG. 7B, the language model may generate a block in which neither the first character string nor the second character string is detected. Alternatively, as in block 304[3] shown in FIG. 7B, the language model may generate a block that includes a third character string detected as a separate typographical error in addition to the second character string.

[0148] In Figure 7B, the second character string is hatched 311b, and the third character string is hatched 311c. Also in Figure 7B, text 323c explaining the third type of error is displayed outside the table including the serial number 326 and the block 327, along with the hatching applied to the third character string.

[0149] [Step S17] In step S17, the processing unit 130 evaluates the acquired block.

[0150] The more blocks a language model generates, the greater the risk of generating sentences that are not intended by the user. Furthermore, because language models are large-scale models, it is difficult to fine-tune them specifically for individual users or specific categories of documents.

[0151] Therefore, it is preferable that the processing unit 130 evaluates the acquired block to determine whether it is appropriate as a sentence or document. For example, appropriate sentence endings or appropriate phrases vary depending on the classification of the sentence or document. Therefore, it is preferable to evaluate the acquired block based on the expression, format, or form of the sentence.

[0152] It is preferable to use an evaluation model for the evaluation. The evaluation model may use a language model and appropriate prompts, but it is more preferable to use a model optimized by machine learning or a rule-based model. By using an optimized model or a rule-based model, high accuracy can be achieved with less computational effort (fewer parameters) than a language model. It is preferable that the model is trained so that sentences or documents that are generally considered favorable in a specific category of documents are highly evaluated. Examples of document categories include official documents and conversational texts. When a model optimized by machine learning is used as the evaluation model, the evaluation model can be said to be a different type of model from a language model.

[0153] For example, if document 301 is a patent application document, the evaluation model is preferably trained to classify the acquired blocks into sentences or documents that are considered preferable for patent application documents and sentences or documents that are difficult to consider preferable for patent application documents. Alternatively, the evaluation model is preferably trained to give a higher score to sentences or documents that are considered more preferable for patent application documents. In the information processing method of this embodiment, the evaluation model is trained so that the score of a block is higher for sentences or documents that are considered more preferable for documents of a specific classification in general.

[0154] In FIG. 8A, a table including a score 328 in addition to a serial number 326 and a block 327 is displayed on the display unit of the information terminal 20.

[0155] The blocks 304 are preferably displayed in descending order of their scores. In Fig. 8A, the blocks 304 are displayed so that the higher the score, the higher the position. In Fig. 8A, as an example, the score of block 304[1] is 0.5, the score of block 304[2] is 0.9, and the score of block 304[3] is 0.2. Therefore, in Fig. 8A, blocks 304[2], 304[1], and 304[3] are displayed in descending order from top to bottom.

[0156] By evaluating the obtained blocks, the user can easily select blocks suitable for the document 301. In addition, the user can select suggested revisions according to the user's preferences or the classification of the document 301.

[0157] Note that blocks with low scores may be configured not to be displayed on the display unit of the information terminal 20. By configuring the blocks with low scores not to be displayed, the number of blocks for which the user checks the evaluation can be reduced, thereby reducing the workload of the user.

[0158] The user selects one of at least one blocks 304 displayed on the display unit of the information terminal 20. Specifically, the user selects a block to be replaced with block 302[1] of document 301. In the information processing method of this embodiment, it is assumed that the user selects block 304[2].

[0159] [Step S18] In step S18, the processing unit 130 updates the document 301.

[0160] Specifically, the processing unit 130 replaces block 302[1] with block 304 selected by the user. Fig. 8B shows a case where block 302[1] is replaced with block 304[2]. As shown in Fig. 8B, by changing the hatching applied to paragraph 0001, the user can easily determine that the correction of paragraph 0001 has been completed.

[0161] This completes the correction of block 302[1]. Similar processing is performed for blocks 302[2] through 302[n]. This completes the update of document 301.

[0162] In some cases, a character string determined to require correction may not be detected in at least one of blocks 302[1] through 302[n]. In such cases, steps S14 through S18 can be omitted for that block.

[0163] [Step S19] In step S19, the output unit 140 outputs the document data. The document data is the document data of the updated document 301, and corresponds to the data DB2 described above.

[0164] As described above, it is possible to correct documents using an external service that provides a language model, and to manage document correction work using the language model.

[0165] <Configuration Example 2 of Information Processing System> Fig. 9 shows another example of the information processing system of this embodiment. The information processing system of this embodiment has an information processing device 11, an information processing device 15, an information terminal 20, and an information processing device 40. In the information processing system shown in Fig. 9, the information processing device 11 is connected to the information processing device 40 via a network 30. The information processing device 11 is also connected to the information processing device 15 and multiple information terminals 20 via a network 31. In Fig. 9, information terminals 20a and 20b are shown as examples of the information terminals 20.

[0166] 9 is mainly used to relay transmission and reception of data between the information terminal 20 and the information processing device 40. Then, in the information processing device 15, data DA2 is created using the data DA1.

[0167] First, the user inputs desired data DA1 into the information processing device 11 using the information terminal 20. The data DA1 is transmitted from the information terminal 20 to the information processing device 11 via the network 31.

[0168] Next, the information processing device 11 transmits the data DA1 to the information processing device 15 via the network 31.

[0169] Next, the information processing device 15 creates data DA2 using the data DA1. The information processing device 15 is configured to execute the various processes performed by the information processing device 10 described above.

[0170] The data DA2 can be created by the information processing device 11. For example, the information processing device 15 may determine whether or not a block needs to be corrected, label the block, and evaluate the block, and the information processing device 11 may create a correction prompt.

[0171] Next, the information processing device 15 transmits the data DA2 to the information processing device 11 via the network 31.

[0172] Next, the information processing device 11 transmits the data DA2 to the information processing device 40 via the network 30.

[0173] Next, the data DB1 is transmitted from the information processing device 40 to the information processing device 11 via the network 30. The data DB1 is text data equivalent to a response sentence corresponding to the data DA2 input to the information processing device 40.

[0174] Next, the information processing device 11 transmits the data DB2 to the information terminal 20 via the network 31. The data DB2 is data based on the data DB1 input to the information processing device 11. As a result, the user can obtain the data DB2, which is the data DA1 updated based on the data DB1.

[0175] The data DB2 can be created by the information processing device 11 or the information processing device 15. For example, the information processing device 11 may input the data DB1 input from the information processing device 40 to the information processing device 15. Then, the information processing device 15 may create the data DB2 based on the data DB1.

[0176] The information processing system shown in FIG. 9, like the information processing system shown in FIG. 1, is suitable for a case where a user who uses the information terminal 20 and a service provider belong to the same organization such as a company.

[0177] For example, it is preferable that data be transmitted and received between the information terminal 20 and the information processing device 11 using a network 31 established within an organization such as a company. It is also preferable that data be transmitted and received between the information processing device 15 and the information processing device 11 using a network 31 established within an organization such as a company. It is also preferable that data be transmitted and received between the information processing device 11 and the information processing device 40 using the Internet. This allows for more secure use of applications, tools, services, etc. provided outside the organization than when the information terminal 20 is directly connected to the information processing device 40 via the Internet. However, connecting the information processing device 15 to the Internet may raise the risk of external cyberattacks, etc. The information processing system shown in FIG. 9 is preferable from a security perspective because it does not require the information processing device 15 to be directly connected to the information processing device 40 via the Internet.

[0178] Fig. 10 is a block diagram of the information processing system shown in Fig. 9. In Fig. 10, an example of the configuration of the information processing device 11 and the information processing device 15 will be mainly described.

[0179] The information processing device 11 includes a receiving unit 110 , a storage unit 120 , a processing unit 130 , an output unit 140 , and a transmission path 150 .

[0180] For the reception unit 110, storage unit 120, output unit 140, and transmission path 150 of the information processing device 11, the descriptions of the information processing device 10 can be referred to. The following mainly describes the differences from the information processing device 10.

[0181] The information processing device 15 includes a receiving unit 115 , a storage unit 125 , a processing unit 135 , an output unit 145 , and a transmission path 155 .

[0182] The receiving unit 110 receives data from the output unit 145 via the network 31 .

[0183] The receiving unit 115 receives data from the output unit 140 via the network 31 .

[0184] The information supplied to the reception unit 115 is supplied to one or both of the storage unit 125 and the processing unit 135 via the transmission path 155 .

[0185] The storage unit 125 has a function of storing a program executed by the processing unit 135. The storage unit 125 may also have a function of storing data created by the processing unit 135 (e.g., calculation results, analysis results, inference results), data input to the reception unit 115, etc. The storage unit 125 may have various configurations applicable to the storage unit 120.

[0186] The processing unit 135 has a function of performing processes such as calculation, analysis, and inference using data supplied from one or both of the reception unit 115 and the storage unit 125. The processing unit 135 can supply the created data (e.g., calculation results, analysis results, and inference results) to one or both of the storage unit 125 and the output unit 145. The processing unit 135 can use various configurations applicable to the processing unit 130.

[0187] The processing unit 135 has a function of acquiring data from the storage unit 125 .

[0188] The processing unit 135 uses the data received by the receiving unit 115 to determine whether correction is necessary and to perform labeling or evaluation.

[0189] The information processing device 15 preferably uses AI for at least a part of the processing, and preferably uses a neural network.

[0190] The output unit 140 outputs the data to the reception unit 115 via the network 31 .

[0191] The output unit 145 outputs information based on the processing result of the processing unit 135. The output unit 145 can supply at least one of the calculation result, analysis result, and inference result of the processing unit 135 to an external device of the information processing device 15.

[0192] The output unit 145 outputs data to the reception unit 110 of the information processing device 10 via the network 31 .

[0193] The transmission path 155 has a function of transmitting data. Data can be transmitted and received between the reception unit 115, the storage unit 125, the processing unit 135, and the output unit 145 via the transmission path 155.

[0194] The processing section 130 and the processing section 135 can each include one or both of a transistor having a metal oxide in a channel formation region (OS transistor) and a transistor having silicon in a channel formation region (Si transistor).

[0195] <Configuration Example 3 of Information Processing System> Fig. 11 shows an example of an information processing system according to this embodiment. In the information processing system shown in Fig. 11, an information processing device 10 is connected to an information processing device 40 and a plurality of information terminals 20 via a network 30. In Fig. 11, the above-mentioned information terminals 20a to 20d are shown as examples of the information terminals 20.

[0196] The flow of data in the information processing system shown in FIG. 11 can be understood by referring to the description of FIG. 2 by replacing the network 31 with the network 30.

[0197] A user of the information processing system can access the information processing device 10 from an information terminal 20. The user can then receive a service that uses the information processing method according to one aspect of the present invention.

[0198] The information processing system shown in FIG. 11 is suitable for a case where a user who uses the information terminal 20 and a service provider belong to different organizations such as companies.

[0199] The information processing device 10 is connected to an information terminal 20 used by a user who receives a service via a network 30 (typically, the Internet). A service provider can provide the service using the information processing device 10. Data is also transmitted and received between the information processing device 10 and the information processing device 40 using the network 30 (typically, the Internet). This allows applications, tools, services, etc. using the information processing device 40 to be used more safely than when the information terminal 20 is directly connected to the information processing device 40 via the Internet.

[0200] <Configuration Example 4 of Information Processing System> Fig. 12 shows an example of an information processing system according to this embodiment. In the information processing system shown in Fig. 12, an information processing device 11 is connected to an information processing device 40 and a plurality of information terminals 20 via a network 30. In Fig. 12, the above-mentioned information terminals 20a to 20d are shown as examples of the information terminals 20. Furthermore, the information processing device 11 is connected to an information processing device 15 via a network 31.

[0201] A user of the information processing system can access the information processing device 11 from an information terminal 20. The user can then receive services using the information processing method according to one aspect of the present invention.

[0202] The information processing system shown in FIG. 12, like the information system shown in FIG. 7, is suitable for a case where a user who uses information terminal 20 and a service provider belong to different organizations such as companies.

[0203] The information processing device 11 is connected to an information terminal 20 used by a user receiving a service via the Internet. Preferably, data is transmitted and received between the information processing device 11 and the information processing device 15 using a network 31 established within an organization such as a company that provides the service. The service provider can provide the service using the information processing device 11 and the information processing device 15. Data is also transmitted and received between the information processing device 11 and the information processing device 40 using the Internet. This allows applications, tools, services, etc. using the information processing device 40 to be used more safely than when the information terminal 20 is directly connected to the information processing device 40 via the Internet. Furthermore, the information processing device 15 does not need to be directly connected to the information processing device 40 via the Internet, which is preferable from the perspective of security.

[0204] According to one aspect of the present invention, it is possible to provide an information processing system or an information processing method that utilizes a language model to support document correction. According to one aspect of the present invention, it is possible to provide a highly convenient information processing system or an information processing method. According to one aspect of the present invention, it is possible to provide a novel information processing system or an information processing method.

[0205] A plurality of configuration examples shown in this embodiment mode can be combined as appropriate.

[0206] 10: information processing device, 11: information processing device, 15: information processing device, 20: information terminal, 20a: information terminal, 20b: information terminal, 20c: information terminal, 20d: information terminal, 21: housing, 30: network, 31: network, 40: information processing device, 110: reception unit, 115: reception unit, 120: storage unit, 125: storage unit, 130: processing unit, 135: processing unit, 140: output unit, 145: output unit, 150: transmission path, 155: transmission path, 210: decision model, 210a: model, 210b: model, 210c: model, 250: language model, 301 : Document, 302[1]: Block, 302[2]: Block, 302[n]: Block, 302: Block, 303: Prompt, 304[1]: Block, 304[2]: Block, 304[3]: Block, 304: Block, 311a: Hatch, 311b: Hatch, 311c: Hatch, 312: Hatch, 321: Paragraph, 322: Block, 323a: Text, 323b: Text, 323c: Text, 324: Mouse pointer, 325: Text, 326: Serial number, 327: Block, 328: Score,

Claims

A reception unit and a processing unit are included. the receiving unit has a function of receiving document data, The processing unit includes: A process of dividing a document included in the document data into a plurality of first blocks; creating a prompt including a target sentence including one of the first blocks and a directive including an instruction to modify the sentence; transmitting the prompt to a language model via a network to obtain at least one second block that is a revision proposal for the target sentence included in the prompt; updating the document data by replacing one of the plurality of first blocks with at least one of the second blocks; A process of determining whether or not correction is required for one or both of the first block and the second block; and performing an evaluation of the second block based on an expression, form, or morphology of the sentence. 　 In claim 1, An information processing system, wherein the document data is a specification belonging to a patent application. 　 In claim 1, An output unit is provided. The output unit has a function of outputting the updated document data. 　 In claim 1, The determination of whether or not the correction is necessary is performed using a determination model, The processing unit is configured to execute a process using the judgment model, An information processing system, wherein the determination model is a different type of model from the language model. 　 In claim 1, The evaluation is performed using an evaluation model; The processing unit is configured to execute a process using the evaluation model, An information processing system, wherein the evaluation model is a different type of model from the language model. 　 In claim 1, A storage unit is provided. The storage unit has a function of storing an instruction sentence in the prompt. 　 Accepts document data, Dividing a document included in the document data into a plurality of first blocks; determining whether or not correction is required for one of the plurality of first blocks; creating a prompt including a target sentence including one of the plurality of first blocks and a directive including an instruction to modify the sentence; transmitting the prompt to a language model via a network to obtain at least one second block that is a revision proposal for the target sentence included in the prompt; evaluating each of the at least one second block based on the expression, style, or form of the sentence; The information processing method includes: updating the document data by replacing one of the plurality of first blocks with at least one of the second blocks. 　 In claim 7, An information processing method comprising: determining whether or not the correction is required for each of at least one of the second blocks. 　 In claim 7, An information processing method, wherein the document data is a specification belonging to a patent application. 　 In claim 7, The determination of whether or not the correction is necessary is performed using a determination model, An information processing method, wherein the determination model is a different type of model from the language model. 　 In claim 7, The evaluation is performed using an evaluation model; An information processing method, wherein the evaluation model is a different type of model from the language model.

Images