Embedded attributes for modifying behaviors of generative AI systems

Abstract

Systems and methods for directing behaviors of a generative artificial intelligence (AI) system are provided. Specifically, a computing device may obtain, for one or more generative artificial intelligence (AI) systems, an input cue associated with a requested task, obtain one or more attributes based on the input cue, modify the input cue based on the one or more embedded attributes, and modify the one or more attributes based on the modified input cue. And providing the modified input cue to one or more generative AI systems.

Description

Background Technology

[0001] Generative artificial intelligence (AI) prompts are instructions used to guide a generative AI system (e.g., a large language model) to complete a requested task. Typically, generative AI behavior is sensitive to input prompts. Therefore, prompts are usually designed to contain relevant information to guide the generative output. This information may include examples (e.g., for few-shot learning), guidance for completing the task (e.g., thought chain guidance), and / or relevant context for completing the task (e.g., relevant documents or paragraphs). Prompts can be a single word, a list of words, one or more phrases, or one or more sentences.

[0002] It is based on these considerations, as well as other general considerations, that the various aspects disclosed herein have been formed. Furthermore, while relatively specific problems may be discussed, it should be understood that the examples should not be limited to solving specific problems identified in the background or elsewhere in this disclosure. Summary of the Invention

[0003] According to examples in this disclosure, an AI guidance system guides the behavior of a generative AI system to complete one or more requested tasks (i.e., generative outputs) based on input prompts. To this end, the AI ​​guidance system generates additional prompts or modifies input prompts based on one or more embedded attributes associated with one or more applications, documents, interfaces, and / or content designed to communicate with the generative AI system based on the input prompts. The embedded attributes of the applications, documents, interfaces, and / or content can be provided by the developers of the applications or interfaces and / or the authors of the documents and content. In other words, the behavior of the generative AI system can be customized or guided at a more granular level by users (e.g., developers and / or authors) using embedded attributes (e.g., by enabling different prompt enhancements for any document or user interface element). It should be understood that utilizing embedded attributes that can supplement input prompts can reduce the workload of current practices that involve writing lengthy, custom prompts to reflect context. It should also be understood that embedded attributes can guide the behavior of the generative AI system on the client or server side.

[0004] According to at least one example of this disclosure, a method for guiding the behavior of a generative artificial intelligence (AI) system is provided. The method may include obtaining an input cue word associated with a requested task for one or more generative AI systems; obtaining one or more attributes based on the input cue word; modifying the input cue word based on the one or more embedded attributes; and providing the modified input cue word to one or more generative AI systems.

[0005] According to at least one example of this disclosure, a method for guiding the behavior of a generative artificial intelligence (AI) system is provided. The method may include obtaining input prompts associated with a requested task for one or more generative AI systems, obtaining one or more attributes based on the input prompts, generating supplementary prompts based on the one or more attributes in response to determining the presence of the one or more attributes, and providing the supplementary prompts and input prompts to the one or more generative AI systems.

[0006] According to at least one example of this disclosure, a computing device is provided for guiding the behavior of a generative artificial intelligence (AI) system. The computing device may include a processor and a memory having a plurality of instructions stored thereon, which, when executed by the processor, cause the computing device to obtain, for one or more generative AI systems, an input cue word associated with a requested task; obtain one or more attributes based on the input cue word; modify the input cue word based on the one or more embedded attributes; and provide the modified input cue word to the one or more generative AI systems.

[0007] Any combination of the above one or more aspects and any other aspect of those one or more aspects. Any one of the one or more aspects as described herein.

[0008] This summary is provided to introduce the selection of concepts in a simplified form, which will be further described in the detailed description below. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to limit the scope of the claimed subject matter. Additional aspects, features, and / or advantages of the examples will be set forth in part in the description which follows, and will be apparent in part from the description, or may be learned by practice of this disclosure. Attached Figure Description

[0009] Non-restrictive and non-exhaustive examples are described with reference to the following figures.

[0010] Figure 1 A block diagram depicts an example of an operating environment in which an AI-guided system can be implemented, according to an example of this disclosure.

[0011] Figure 2 A flowchart depicts an example method for guiding the behavior of one or more generative AI systems according to examples of this disclosure.

[0012] Figure 3A Example meta prompts in JavaScript are shown as examples of how to guide one or more generative AI systems to fill out forms on a website, according to this disclosure.

[0013] Figure 3BA screenshot of an exemplary webpage is depicted, in which user prompts are provided to one or more generative AI systems for requesting the writing of a summary of a sample item.

[0014] Figure 3C It describes the attributes, including those used to guide one or more generative AI systems in writing sample projects. Figure 3B The image shows a screenshot of the HTML code for the webpage.

[0015] Figure 4A and Figure 4B An overview of example generative machine learning models that can be used based on the examples in this disclosure is presented.

[0016] Figure 5 This is a block diagram illustrating example physical components of a computing device in which aspects of this disclosure can be utilized.

[0017] Figure 6 This is a simplified block diagram of a computing device in which various aspects of this disclosure can be utilized.

[0018] Figure 7 This is a simplified block diagram of a distributed computing system in which various aspects of the present disclosure can be put into practice. Detailed Implementation

[0019] In the following detailed description, reference is made to the accompanying drawings, which form a part of the description, and specific aspects or examples are illustrated therein. These aspects may be combined, other aspects may be utilized, and structural changes may be made without departing from this disclosure. The aspects may be practiced as methods, systems, or devices. Accordingly, the aspects may take the form of hardware implementations, entirely software implementations, or implementations combining software and hardware aspects. Therefore, the following detailed description should not be considered limiting, and the scope of this disclosure is defined by the appended claims and their equivalents.

[0020] Generative artificial intelligence (AI) prompts are instructions used to guide a generative AI system (e.g., a large language model) to complete a requested task. Typically, generative AI behavior is sensitive to input prompts. Therefore, prompts are usually designed to contain relevant information to guide the generative output. This information may include examples (e.g., for few-shot learning), guidance for completing the task (e.g., thought chain guidance), and / or relevant context for completing the task (e.g., relevant documents or paragraphs). Prompts can be a single word, a list of words, one or more phrases, or one or more sentences.

[0021] Current approaches to guiding generative AI behavior typically involve generating or enriching cue words for the generative AI system. For example, users can generate free-form cue words to guide AI behavior to perform a requested task. Alternatively or additionally, cue words can be automatically enriched with the required data before requesting generative output. In some examples, some text of the cue words can be hard-coded to guide interaction (e.g., cue words used for sentiment analysis or generalization). However, guiding generative AI behavior tailored to specific content or applications can be challenging for users and developers.

[0022] According to examples in this disclosure, an AI guidance system guides the behavior of a generative AI system to complete one or more requested tasks (i.e., generative outputs) based on input prompts. To this end, the AI ​​guidance system generates additional prompts or modifies input prompts based on one or more embedded attributes associated with one or more applications, documents, interfaces, and / or content designed to communicate with the generative AI system based on the input prompts. The embedded attributes of the applications, documents, interfaces, and / or content can be provided by the developers of the applications or interfaces and / or the authors of the documents and content. For example, embedded attributes can be labels, prompts, tags, or any indications that provide additional instructions to the generative AI system. In other words, the behavior of the generative AI system can be customized or guided at a more granular level by users (e.g., developers and / or authors) using embedded attributes (e.g., by enabling different prompt enhancements for any document or user interface element). For example, embedded attributes can be received via developer tools used to create documents, applications, web pages, etc., productivity applications, or another application. It should be understood that embedded attributes are extensible. In other words, by leveraging embedding attributes that can supplement input prompts, the current practice of writing lengthy, custom prompts to reflect context can be reduced. It should also be understood that embedding attributes can guide the behavior of generative AI systems on either the client or server side.

[0023] Figure 1 A block diagram depicts an example of an operating environment 100 in which an AI bootstrapping system can be implemented according to an example of this disclosure. For this purpose, the operating environment 100 includes a computing device 120 associated with a user 110. The operating environment 100 may also include one or more remote devices, such as an AI platform server 150 and a bootstrapping platform server 130, communicatively coupled to the computing device 120 via a network 170. The network 170 may include any kind of computing network, including but not limited to wired or wireless local area networks (LANs), wired or wireless wide area networks (WANs), and / or the Internet.

[0024] AI platform server 150 includes one or more generative AI systems 160 and is configured to render one or more generative AI systems 160. The generative AI system 160 may include generative large language machine learning models, transformer models, other types of machine learning models, or combinations thereof. Computing device 120 has a processor 122, memory 124, and a communication interface 126. Computing device 120 may be, but is not limited to, a computer, laptop, mobile device, smartphone, tablet, portable device, or any other suitable computing device capable of communicating with one or more generative AI systems 160. It should be understood that, in some aspects, computing device 120 may execute one or more generative AI systems 160.

[0025] The guidance platform server 130 includes an AI guidance system 140 configured to communicate with the computing device 120 and the AI ​​platform server 150. However, it should be understood that, in some aspects, the AI ​​guidance system 140 may be executed on the AI ​​platform server 150 and / or the computing device 120. The AI ​​guidance system 140 is also configured to guide the behavior of one or more generative AI systems to complete one or more requested tasks (i.e., generative outputs) based on the input prompts by supplementing or modifying input prompts for one or more generative AI systems. Specifically, the AI ​​guidance system 140 is configured to generate additional prompts or modify input prompts based on one or more embedded attributes associated with one or more applications, documents, interfaces, and / or content designed to communicate with the generative AI system based on the input prompts.

[0026] It should be understood that embedding attributes do not need to be human-interpretable. For example, embedding attributes can simply provide an embedding vector, or refer to some point or direction in the latent space of an AI model. This allows embedding attributes to convey concepts that are difficult to put into words (e.g., style cues). In some aspects, embedding attributes can be keys that can be retrieved from memory, so that proprietary cue word information is not revealed to the user. Embedding attributes can evolve into broader internet communication protocols that support the use of large language model (LLM) agents (e.g., IoT devices) representing user actions to solve certain tasks. To this end, the AI ​​guidance system 140 includes a cue word receiver 142, an attribute determiner 144, a cue word manager 146, and a cue word provider 148.

[0027] The prompt word acquirer 142 is configured to receive, acquire, or otherwise acquire prompt words for one or more generative AI systems 160. For example, user 110 may provide prompt words for one or more generative AI systems 160. A prompt word is an input or query provided by user 110 or a program to one or more generative AI systems 160 to elicit a requested output or response from one or more generative AI systems 160. The prompt word describes the requested task to be performed by one or more generative AI systems 160. A prompt word can be a natural language sentence or question, or a code snippet or command, or any combination of text or code.

[0028] Attribute determiner 144 is configured to determine one or more applications, documents, interfaces, and / or content configured to communicate with a generative AI system to perform a requested task based on prompts. Attribute determiner 144 is also configured to determine whether one or more embedded attributes are associated with one or more applications, documents, interfaces, and / or content. As described above, embedded attributes of applications, documents, interfaces, and / or content can be provided by the developers of the applications or interfaces and / or the authors of the documents and content. Users (e.g., developers or authors) can use embedded attributes to enable different prompt enhancements for any document or user interface element. For example, an embedded attribute can be a label, hint, tag, or any indication that provides additional instructions to the generative AI system. It should be understood that utilizing embedded attributes that can supplement input prompts can reduce the workload of current practices that require writing lengthy, custom prompts to reflect context. It should also be understood that embedded attributes can guide the behavior of the generative AI system on the client or server side.

[0029] The prompt manager 146 is configured to generate new supplementary prompts and / or modify the original prompts input by the user for one or more generative AI systems based on one or more embedding attributes.

[0030] The cue word provider 148 is configured to provide one or more cue words to one or more generative AI systems. The one or more cue words may include new supplementary cue words, modified cue words, and / or the original input cue words. For example, the cue word provider 148 is configured to provide new supplementary cue words to one or more generative AI systems along with the original input cue words.

[0031] Now for reference Figure 2 This disclosure provides a method 200 for guiding the behavior of one or more generative AI systems, according to examples. The general sequence of the steps of method 200 is as follows: Figure 2 The method is shown in the diagram. Typically, method 200 begins at 202 and ends at 216. Method 200 may include more or fewer steps, or may be combined with... Figure 2The steps shown are arranged differently in different order. For illustrative purposes, method 200 is performed by a server (e.g., bootstrapping platform server 130). However, it should be understood that one or more steps of method 200 may be performed by another device (e.g., AI platform server 150 and / or computing device 120).

[0032] Specifically, in some aspects, method 200 can be executed by an AI bootstrapping system (e.g., 140) executed on bootstrapping platform server 130. For example, AI bootstrapping system 140 can be any tool capable of generating new prompts or modifying existing prompts and communicatively coupled to a computing device (e.g., computing device 120) providing input prompts and one or more generative systems 160. For example, bootstrapping platform server 130 can be any suitable computing device capable of communicating with computing device 120. For example, computing device 120 can be, but is not limited to, a computer, notebook, laptop, mobile device, smartphone, tablet, portable device, or any other suitable computing device capable of communicating with one or more generative AI systems (e.g., 160). Method 200 can be executed as a set of computer-executable instructions executed by a computer system and encoded or stored on a computer-readable medium. Furthermore, method 200 can be executed by gates or circuits associated with a processor, application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), system-on-a-chip (SOC), or other hardware device. Hereinafter, method 200 will be referred to in conjunction with... Figure 1 and Figure 4 to Figure 7 The described systems, components, modules, software, data structures, user interfaces, etc., are explained.

[0033] Method 200 begins at operation 202, where the process can proceed to 204. At operation 204, the AI ​​guidance system 140 receives input prompts for one or more generative AI systems 160. The input prompts are inputs or queries provided by a user or program to one or more generative AI systems 160 to elicit a requested output or response from the one or more generative AI systems. The input prompts describe the requested task to be performed by the one or more generative AI systems 160. As described above, the input prompts can be natural language sentences or questions, code snippets or commands, or any combination of text or code.

[0034] At operation 206, the AI ​​guidance system 140 identifies one or more applications, documents, interfaces, and / or content associated with the input prompt. For example, the AI ​​guidance system 140 identifies one or more applications, documents, interfaces, and / or content intended to be used to perform the requested output based on the input prompt.

[0035] At operation 208, AI guidance system 140 determines whether one or more embedded attributes are associated with one or more applications, documents, interfaces, and / or content. If AI guidance system 140 determines at operation 210 that one or more embedded attributes exist, method 200 proceeds to operation 212.

[0036] At operation 212, the AI ​​guidance system 140 generates new supplementary prompts and / or modifies input prompts based on one or more attributes.

[0037] At operation 214, the AI ​​guidance system 140 provides one or more prompts to one or more generative AI systems. The one or more prompts may include new supplementary prompts, modified prompts, and / or the original input prompts. It should be understood that new supplementary prompts are provided to one or more generative AI systems along with the original input prompts.

[0038] Returning to reference operation 210, if the AI ​​guidance system 140 determines at operation 210 that the embedded attribute does not exist, then method 200 jumps forward to operation 214 to provide raw input prompts to one or more generative AI systems.

[0039] For example, consider an email client that uses a Large Language Model (LLM) to help users compose emails (e.g., autocomplete or smart replies). The email application developer can include embedded attributes around the HTML text area of ​​the email body, which guides the AI ​​to compose the email in the same tone as other emails previously sent to a specific recipient. In such an example, a user can provide input prompts to a Large Language Model (LLM) AI to send an email to a recipient via the email application. The AI ​​guidance system receives the input prompts and determines the embedded attributes around the HTML text area of ​​the email body included in the email application. Based on the embedded attributes, the AI ​​guidance system can generate supplementary prompts and provide the original input prompts to the LLM AI along with the supplementary prompts. Additionally or alternatively, the AI ​​guidance system can modify the input prompts based on the embedded attributes and provide the modified input prompts to the LLM AI. Based on the given prompts(s), the LLM AI will generate an email to the recipient in the same tone as other emails previously sent to the user's recipient.

[0040] In other examples, internet forums or social media sites may post one or more embedded attributes to guide the tone of discussion or provide writing assistance, ensuring that it does not violate forum rules or guidelines (e.g., forums often have rules for self-promotion or off-topic discussions). In such an example, a user could provide input prompts to a Large Language Model (LLM) AI to generate a post with a recent photo and relevant description on a social media site. The AI ​​guidance system receives the input prompts and determines that the social media site includes embedded attributes indicating rules and restrictions regarding photo and text size. Based on these embedded attributes, the AI ​​guidance system can generate supplementary prompts and provide the original input prompts to the LLM AI along with the supplementary prompts. Additionally or alternatively, the AI ​​guidance system can modify the input prompts based on the embedded attributes and provide the modified input prompts to the LLM AI. Based on the given prompts(s), the LLM AI will generate a post for the user with a recent photo of a specific size and a description that follows rules (e.g., using hashtags).

[0041] In other examples, consider a webpage containing free-form text fields (e.g., application-specific, compliance-specific, and legal-specific) designed to gather responses from users. The webpage author may include different embedding attributes around each HTML text area, containing examples of appropriate responses that can be used as guidance for the AI. For example, a job application webpage might contain free-form text fields for applicants to write short descriptions of their suitability for the position, and text fields for applicants to write short descriptions of their relevant experience. The embedding attributes around the text areas may include examples of appropriate responses. In such an example, a user could provide input prompts to a Large Language Model (LLM) AI to populate text fields in a specific job application webpage. In response, the AI ​​guidance system receives the input prompts and determines that the job application webpage includes embedding attributes around the text areas. Based on these embedding attributes, the AI ​​guidance system can generate supplementary prompts and provide the original input prompts to the LLM AI along with the supplementary prompts. Additionally or alternatively, the AI ​​guidance system can modify the input prompts based on the embedding attributes and provide the modified input prompts to the LLM AI. Based on the given prompts(s), the LLM AI will generate populates for each text field according to the user's requirements and the format described in the embedding attributes.

[0042] Figure 3AMeta-prompt 300 in JavaScript is provided for the above-described text field example according to the present disclosure. Meta-prompt 300 is configured to guide a Large Language Model (LLM) AI to fill in forms on a website based on a “userPrompt” (i.e., an input prompt provided by the user), as shown in 304. In meta-prompt 300, the LLM AI determines, for example, whether text appears near the form field, whether the form field has a label, and whether the user highlights any existing text used to fill in the requested form. Additionally, the LLM AI determines whether the form field being edited provides an “aiAttribute” (i.e., one or more embedded attributes around the form field), as shown in 302. Based on user prompt 304.

[0043] Figure 3B and Figure 3C The example webpage shown has an embedding attribute used to guide one or more generative AI systems to perform a requested task based on user prompts. For example, a user could access a webpage for submitting projects on the EthicsReview Portal. Figure 3B As shown in the screenshot, webpage 310 may include a populateable text area 312 for providing a brief description of the project and its objectives. Users can utilize one or more generative AI systems to fill out the form. For this purpose, users can provide user prompts to one or more generative AI systems to “write a summary of the sample project.” Additionally, the authors of the Ethics Review Portal webpage 310 can include one or more embedding attributes around the HTML text area on the webpage to guide one or more generative AI systems. For example, as... Figure 3C As depicted in screenshot 320 of the HTML (Hypertext Markup Language) code of the webpage, attribute 322 can be embedded in the HTML text area surrounding the project description field 312. This attribute 322 indicates that "the project description should be written for a general audience. Avoid overly technical terms, acronyms, or internal project names."

[0044] Figure 4A and Figure 4B An overview of example generative machine learning models that can be used based on the aspects described in this paper is presented. First, refer to... Figure 4A Conceptual diagram 400 depicts an overview of a pre-trained generative model package 404 according to the aspects described herein, which processes input 402 to generate model output for storing entries in generative model output 406 and / or retrieving information (e.g., suggestions and / or suggested modifications) from generative model output 406.

[0045] In the example, the generative model package 404 is pre-trained based on various inputs (e.g., various human languages, various programming languages, and / or various content types) and therefore does not need to be fine-tuned or trained for a specific scenario. Instead, the generative model package 404 can be pre-trained more generally such that the input 402 includes cue words that are generated, selected, or designed to induce the generative model package 404 to produce certain generative model outputs 406. It should be understood that the input 402 and the generative model output 406 can each include any type of content from various content types, including but not limited to text output, image output, audio output, video output, programming output, and / or binary output. In the example, the input 402 and the generative model output 406 can have different content types, as could be the case when the generative model package 404 includes a generative multimodal machine learning model.

[0046] Thus, generative model package 404 can be used in any of a variety of scenarios, and furthermore, different generative model packages can be used to replace generative model package 404 with minimal modification to other related aspects (e.g., similar to those discussed in this paper). Figure 1 (To those aspects described in Figure 3). Accordingly, the generative model package 404 operates as a tool used for machine learning processing, wherein certain inputs 402 of the generative model package 404 are generated programmatically or otherwise determined, thereby enabling the generative model package 404 to produce a model output 406, which can subsequently be used for further processing.

[0047] Generative model package 404 can be provided or otherwise used according to any of the various paradigms. For example, generative model package 404 can be used on computing devices (e.g., Figure 1 The computing device 140 in the system can be used locally, or it can be accessed from a machine learning service (e.g., Figure 1 Server 160 in the middle is accessed. In other examples, aspects of the generative model package 404 are distributed across multiple computing devices. In some instances, the generative model package 404 may be accessed via an application programming interface (API), such as that provided by the operating system of the computing device and / or by machine learning services and other examples.

[0048] Referring now to aspects of the generative model package 404, the generative model package 404 includes input tokenization 408, input embedding 410, model layer 412, output layer 414, and output decoding 416. In the example, input tokenization 408 processes input 402 to generate input embedding 410, which includes a sequence of symbol representations corresponding to input 402. Accordingly, input embedding 410 is processed by model layer 412, output layer 414, and output decoding 416 to produce model output 406. Figure 4B The example architecture corresponding to Generative Model Package 404 is depicted below, and it is discussed in further detail. Even so, it should be understood that the architectures shown and described herein should not be considered limiting, and any of a variety of other architectures may be used in other examples.

[0049] Figure 4B This is a conceptual diagram depicting an example architecture 450 of a pre-trained generative machine learning model that can be used according to the aspects described herein. As mentioned above, any of the various alternative architectures and corresponding ML models can be used in other examples without departing from the aspects described herein.

[0050] As shown in the figure, architecture 450 processes input 402 to produce generative model output 406, aspects of which are discussed above regarding... Figure 4A The architecture 450 is described as a transformer model comprising an encoder 452 and a decoder 454. The encoder 452 processes the input embedding 458 (in aspects which can be similar to...). Figure 4A The input embedding 458 (input embedding 410) includes a sequence of symbolic representations corresponding to input 456. In the example, input 456 includes content data 402, which corresponds to content items.

[0051] Furthermore, positional encoding 460 can incorporate information about the relative and / or absolute positions of the lexical units for the input embedding 458. Similarly, the output embedding 474 includes a sequence of symbolic representations corresponding to the output 472, and positional encoding 476 can similarly incorporate information about the relative and / or absolute positions of the lexical units for the output embedding 474.

[0052] As shown in the figure, encoder 452 includes example layer 470. It should be understood that any number of such layers can be used, and the architecture depicted is simplified for illustrative purposes. Example layer 470 includes two sub-layers: a multi-head attention layer 462 and a feedforward layer 466. In the example, residual connections are included around each layer 462, 466, and normalization layers 464 and 468 are included after layers 462, 466.

[0053] Decoder 454 includes example layer 490. Similar to encoder 452, any number of such layers can be used in other examples, and for illustrative purposes, the architecture of decoder 452 depicted is simplified. As shown, example layer 490 includes three sublayers: a masked multi-head attention layer 478, a multi-head attention layer 482, and a feedforward layer 486. Aspects of multi-head attention layer 482 and feedforward layer 486 can be similar to those discussed above regarding multi-head attention layer 462 and feedforward layer 466, respectively. Additionally, masked multi-head attention layer 478 performs multi-head attention on the output of encoder 452 (e.g., output 472). In the example, masked multi-head attention layer 478 prevents positional attention to subsequent positions. This mask, combined with offset embedding (e.g., offsetting by one position, as shown in multi-head attention layer 482), ensures that the prediction for a given position depends on the known output of one or more positions smaller than that given position. As shown in the figure, residual connections are also included around layers 478, 482 and 486, and normalized layers 480, 484 and 488 are included after layers 478, 482 and 486, respectively.

[0054] Multi-head attention layers 462, 478, and 482 can each use a set of linear projections to linearly project the query, key, and value onto the corresponding dimension. Each linear projection can be processed using an attention function (e.g., dot product or additive attention) to produce an n-dimensional output value for each linear projection. The resulting values ​​can be concatenated and projected again, such that the values ​​are subsequently... Figure 4B The data is processed as shown (e.g., by the corresponding normalization layer 464, 480, or 484).

[0055] Feedforward layers 466 and 486 can each be a fully connected feedforward network applied to each location. In the example, feedforward layers 466 and 486 each include multiple linear transformations with modified linear unit activations between them. In the example, each linear transformation is the same across different locations, while different parameters can be used compared to other linear transformations of the feedforward network.

[0056] Additionally, aspects of the linear transformation 492 can be analogous to the linear transformations discussed above with respect to the multi-head attention layers 462, 478, and 482, and the feedforward layers 466 and 486. The Softmax 494 can also convert the output of the linear transformation 492 into the predicted next lexical probability, as indicated by the output probability 496. It should be understood that the architecture shown is provided as an example, and in other examples, any of various other model architectures can be used, depending on the aspects disclosed.

[0057] Accordingly, the output probability 496 can thus form a generative model 406 based on the aspects described herein, such that the output of the generative ML model (e.g., which may include one or more sentiment embeddings and one or more retrieved content items) is used as input for determining an action based on the aspects described herein. In other examples, the generative model output 406 is provided as output for retrieving one or more previously retrieved content items.

[0058] Figures 5 to 7 The associated description provides a discussion of various operating environments in which the aspects of this disclosure can be practiced. However, regarding Figures 5 to 7 The devices and systems shown and discussed are for illustrative purposes only and are not intended to limit the large number of computing device configurations that can be used to practice the aspects of this disclosure described herein.

[0059] Figure 5 This is a block diagram illustrating the physical components (e.g., hardware) of a computing device 500 that can implement various aspects of this disclosure. The computing device components described below can be adapted to the computing device described above, including those related to machine learning services (e.g., productivity platform server 160), and the above-mentioned... Figure 1 The computing device 140 described. In a basic configuration, the computing device 500 may include at least one processing unit 502 and system memory 504. Depending on the configuration and type of the computing device, the system memory 504 may include, but is not limited to, volatile memory (e.g., random access memory), non-volatile memory (e.g., read-only memory), flash memory, or any combination of these memories.

[0060] System memory 504 may include operating system 505 and one or more program modules 506 adapted to run software application 520, such as one or more components supported by the system described herein. As an example, system memory 504 may store content acquisition manager 521 and / or content retrieval manager 522. Operating system 505 may, for example, be adapted to control the operation of computing device 500.

[0061] Furthermore, aspects of this disclosure can be practiced in conjunction with graphics libraries, other operating systems, or any other applications, and are not limited to any particular application or system. This basic configuration is... Figure 5 The components within the dashed line 508 are shown. The computing device 500 may have additional features or functions. For example, the computing device 500 may also include additional data storage devices (removable and / or non-removable), such as disks, optical discs, or magnetic tapes. Such additional storage... Figure 5 The middle part is shown by removable storage device 509 and non-removable storage device 510.

[0062] As described above, multiple program modules and data files can be stored in system memory 504. When executed on processing unit 502, program module 506 (e.g., application 520) can perform processes including, but not limited to, those described herein. Other program modules that can be used according to various aspects of this disclosure may include email and contact applications, word processing applications, spreadsheet applications, database applications, slideshow applications, drawing or computer-aided applications, etc.

[0063] Furthermore, aspects of this disclosure can be implemented on circuits including discrete electronic components, packaged or integrated electronic chips containing logic gates, circuits utilizing microprocessors, or single chips containing electronic components or microprocessors. For example, aspects of this disclosure can be implemented via a system-on-a-chip (SOC), wherein... Figure 5 Each or many of the components shown can be integrated onto a single integrated circuit. Such a SoC device may include one or more processing units, graphics units, communication units, system virtualization units, and various application functions, all integrated (or “programmed”) onto a chip substrate as a single integrated circuit. When operating via the SoC, the capabilities described herein regarding the client switching protocol can be operated via dedicated logic integrated onto a single integrated circuit (chip) along with other components of the computing device 500. Some aspects of this disclosure can also be practiced using other techniques capable of performing logical operations (e.g., AND, OR, and NOT), including but not limited to mechanical, optical, fluid, and quantum technologies. Furthermore, aspects of this disclosure can be practiced within a general-purpose computer or in any other circuit or system.

[0064] The computing device 500 may also have one or more input devices 512, such as a keyboard, mouse, pen, voice or speech input device, touch or swipe input device, etc. Output devices 514, such as a display, speaker, printer, etc., may also be included. The foregoing devices are examples, and other devices may be used. The computing device 500 may include one or more communication connections 516 that allow communication with other computing devices 550. Examples of suitable communication connections 516 include, but are not limited to, radio frequency (RF) transmitters, receivers, and / or transceiver circuitry; universal serial buses (USB), parallel and / or serial ports.

[0065] As used herein, the term computer-readable medium may include computer storage media. Computer storage media may include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, or program modules. System memory 504, removable storage device 509, and non-removable storage device 510 are examples of computer storage media (e.g., memory storage). Computer storage media may include RAM, ROM, electrically erasable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical storage, magnetic tape cassette, magnetic tape, disk storage or other magnetic storage devices, or any other article of manufacture that can be used to store information and is accessible by computing device 500. Any such computer storage medium may be part of computing device 500. Computer storage media does not include carrier waves or other propagated or modulated data signals.

[0066] Communication media can be embodied in computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and include any information delivery medium. The term "modulated data signal" can describe a signal having one or more characteristics set or altered in a manner that encodes information in the signal. By way of example and not limitation, communication media can include wired media such as wired networks or direct wired connections, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0067] Figure 6 System 600 is illustrated. System 600 can be, for example, a mobile computing device, such as a mobile phone, smartphone, wearable computer (such as a smartwatch), tablet computer, laptop computer, etc., and aspects of this disclosure can be practiced using system 600. In one example, system 600 is implemented as a "smartphone" capable of running one or more applications (e.g., browser, email, calendar, contact manager, messaging client, game, and media client / player). In some aspects, system 1002 is integrated as a computing device, such as an integrated personal digital assistant (PDA) and cordless phone.

[0068] In a basic configuration, such a mobile computing device is a handheld computer with both input and output elements. System 600 typically includes a display 605 and one or more input buttons that allow the user to input information into system 600. The display 605 can also be used as an input device (e.g., a touchscreen display).

[0069] If included, optional side input elements allow for further user input. For example, a side input element could be a rotary switch, a button, or any other type of manual input element. Alternatively, system 600 may include more or fewer input elements. For example, in some aspects, display 605 may not be a touchscreen. In another example, an optional keyboard 635 may also be included, which could be a physical keyboard or a “soft” keyboard generated on a touchscreen display.

[0070] In various aspects, output elements include a display 605 for displaying a graphical user interface (GUI), visual indicators (e.g., light-emitting diodes 620), and / or an audio transducer 625 (e.g., a speaker). In some aspects, a vibration transducer is included to provide tactile feedback to the user. In yet another aspect, input and / or output ports are included, such as audio inputs (e.g., microphone jacks), audio outputs (e.g., headphone jacks), and video outputs (e.g., HDMI ports) for sending or receiving signals from external devices.

[0071] One or more applications 666 may be loaded into memory 662 and run on or associated with operating system 664. Examples of applications include telephone dialers, email programs, personal information management (PIM) programs, word processing programs, spreadsheet programs, internet browser programs, messaging programs, etc. System 602 also includes a non-volatile storage area 668 within memory 662. The non-volatile storage area 668 may be used to store persistent information that should not be lost if system 602 is powered off. Applications 666 may use information and store it in the non-volatile storage area 668, such as emails or other messages used by email applications. A synchronization application (not shown) also resides on system 602 and is programmed to interact with a corresponding synchronization application residing on the host computer to keep the information stored in the non-volatile storage area 668 synchronized with the corresponding information stored on the host computer. It should be understood that other applications may be loaded into memory 662 and run on system 600 as described herein (e.g., content acquisition manager, content retrieval manager, etc.).

[0072] System 602 has a power supply 670, which can be implemented as one or more batteries. The power supply 670 may also include an external power source, such as an AC adapter or a power docking station for replenishing or recharging the batteries.

[0073] System 602 may also include a radio interface layer 672 that performs functions for transmitting and receiving radio frequency communications. Radio interface layer 672 facilitates wireless connectivity between system 602 and the "external world" via a communications operator or service provider. Transmissions to and from radio interface layer 672 are conducted under the control of operating system 664. In other words, communications received by radio interface layer 672 can be propagated to application 666 via operating system 664, and vice versa.

[0074] A visual indicator 620 can be used to provide visual notifications, and / or an audio interface 674 can be used to generate audible notifications via an audio transducer 625. In the illustrated example, the visual indicator 620 is a light-emitting diode (LED), and the audio transducer 625 is a speaker. These devices can be directly coupled to a power supply 670 such that when activated, they remain on for the duration indicated by the notification mechanism, even if the processor 660 and other components may be turned off to conserve battery power. The LED can be programmed to remain on indefinitely until the user takes action to indicate the device's power-on status. The audio interface 674 is used to provide and receive audible signals to and from the user. For example, in addition to being coupled to the audio transducer 625, the audio interface 674 can also be coupled to a microphone to receive audible input, such as to facilitate telephone conversations. According to various aspects of this disclosure, the microphone can also be used as an audio sensor to facilitate control of notifications, as described below. System 602 may also include a video interface 676, which enables the operation of the onboard camera 630 to record still images, video streams, etc.

[0075] It should be understood that system 600 may have additional features or functions. For example, system 600 may also include additional data storage devices (removable and / or non-transferable), such as disks, optical discs, or magnetic tapes. Such additional storage... Figure 6 The non-volatile storage region 668 is shown in the middle.

[0076] As described above, data / information generated or acquired and stored via system 600 can be stored locally, or the data can be stored on any number of storage media, which can be accessed by the device via radio interface layer 672 or via a wired connection between system 600 and a separate computing device associated with system 600 (e.g., a server computer in a distributed computing network such as the Internet). It should be understood that such data / information can be accessed via radio interface layer 672 or via a distributed computing network. Similarly, such data / information can be easily transferred between computing devices for storage and use according to any of the various data / information transmission and storage components (including email and collaborative data / information sharing systems).

[0077] Figure 7 One aspect of the architecture of a system for processing data received at a computing system from a remote source (e.g., a personal computer 704, a tablet computing device 706, or a mobile computing device 708) is illustrated, as described above. Content displayed at server device 702 can be stored in different communication channels or other types of storage. For example, various documents can be stored using a directory service 724, a portal website 725, an email service 726, instant messaging storage 728, or a social networking site 730.

[0078] Application 720 (e.g., similar to application 520) can be employed by a client communicating with server device 702. Additionally or alternatively, content acquisition manager 791 and / or content retrieval manager 792 can be employed by server device 702. Server device 702 can provide data to and from client computing devices such as personal computer 704, tablet computing device 706, and / or mobile computing device 708 (e.g., smartphone) via network 715. As an example, the aforementioned computer system can be embodied in personal computer 704, tablet computing device 706, and / or mobile computing device 708 (e.g., smartphone). In addition to receiving graphics data that can be preprocessed at the graphics originating system or post-processed at the receiving computing system, any of these examples of computing devices can obtain content from storage 716.

[0079] It should be understood that the aspects and functions described herein can operate on distributed systems (e.g., cloud-based computing systems), where application functions, memory, data storage and retrieval, and various processing functions can remotely operate on each other over distributed computing networks (such as the Internet or intranets). Various types of user interfaces and information can be displayed via onboard computing device displays or via remote display units associated with one or more computing devices. For example, various types of user interfaces and information can be projected onto a wall, displaying and interacting with them. Interaction with various aspects of this disclosure using multiple computing systems in which they are practiced includes: key input, touchscreen input, voice or other audio input, gesture input, etc., wherein in gesture input, the associated computing device is equipped with detection (e.g., camera) functions for acquiring and interpreting user gestures used to control the computing device.

[0080] For example, the aspects of this disclosure have been described above with reference to block diagrams and / or operational illustrations of methods, systems, and computer program products according to various aspects of this disclosure. Functions / actions indicated in the blocks may not occur in the order shown in any flowchart. For example, depending on the functions / actions involved, two blocks shown consecutively may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order.

[0081] The description and illustrations of one or more aspects provided in this application are not intended to limit or restrict the scope of this disclosure in any way. The aspects, examples, and details provided in this application are considered sufficient to convey ownership and enable others to make and use the claimed aspects of this disclosure. The claimed disclosure should not be construed as limited to any aspect, example, or detail provided in this application. Whether shown and described in combination or separately, various features (both structural and methodological) are intended to be selectively included or omitted to produce aspects having a particular set of features. Having provided the description and illustrations of this application, those skilled in the art can conceive of variations, modifications, and alternatives falling within the spirit of the broader aspects of the overall inventive concept embodied in this application, without departing from the broader scope of the claimed disclosure.

[0082] Furthermore, the aspects and functions described herein can operate on distributed systems (e.g., cloud-based computing systems), where application functions, memory, data storage and retrieval, and various processing functions can operate remotely to each other on distributed computing networks (such as the Internet or intranets). Various types of user interfaces and information can be displayed via onboard computing device displays or via remote display units associated with one or more computing devices. For example, various types of user interfaces and information can be projected onto a wall, displaying and interacting with them. Interaction with various aspects of this disclosure using multiple computing systems in practice includes: key input, touchscreen input, voice or other audio input, gesture input, etc., wherein in gesture input, the associated computing device is equipped with detection (e.g., camera) functions for acquiring and interpreting user gestures used to control the computing device.

[0083] The phrases “at least one,” “one or more,” “or,” and “and / or” are both combined and separate open-ended expressions in operation. For example, each of the expressions “at least one of A, B, and C,” “at least one of A, B, or C,” “one or more of A, B, and C,” “one or more of A, B, or C,” “A, B, and / or C,” and “A, B, or C” means A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B, and C together.

[0084] The term "a" refers to one or more of the same entity. Therefore, the terms "a," "one or more," and "at least one" are used interchangeably herein. It should also be noted that the terms "comprising," "including," and "having" are used interchangeably.

[0085] As used herein, the term "automatic" and its variations refer to any process or operation, typically continuous or semi-continuous, that requires no substantial human input to perform. However, a process or operation can be automatic even if its execution uses substantial or non-substantial human input, provided that input is received before its execution. Human input is considered substantial if it influences how the process or operation will be performed. Human input that consents to the execution of a process or operation is not considered "substantial."

[0086] Any steps, functions, and operations discussed in this article can be performed continuously and automatically.

[0087] Example systems and methods of this disclosure have been described with respect to computing devices. However, to avoid unnecessarily obscuring this disclosure, several known structures and devices have been omitted from the foregoing description. Such omissions should not be construed as limiting. Specific details have been set forth to provide an understanding of this disclosure. However, it should be understood that this disclosure can be practiced in various ways beyond the specific details set forth herein.

[0088] Furthermore, while the examples shown herein illustrate various components of a co-located system, some components of the system may be located remotely in distant portions of a distributed network (such as a LAN and / or the Internet) or remotely within a dedicated system. Therefore, it should be understood that system components may be combined into one or more devices, such as servers, communication equipment, or may coexist on specific nodes of a distributed network, such as analog and / or digital telecommunications networks, packet-switched networks, or circuit-switched networks. As will be understood from the foregoing description, and for computational efficiency reasons, system components may be positioned anywhere within the distributed network of components without affecting the operation of the system.

[0089] Furthermore, it should be understood that the various links connecting the elements can be wired links or wireless links, or any combination thereof, or any other known or to be developed element(s) capable of supplying data to and / or transmitting data from the connected elements. These wired or wireless links can also be secure links and capable of transmitting encrypted information. The transmission medium used as the link can be, for example, any suitable carrier wave for electrical signals, including coaxial cable, copper wire, and optical fiber, and can take the form of sound waves or light waves, such as those generated during radio wave and infrared data communication.

[0090] Although the flowchart has been discussed and shown with respect to a particular sequence of events, it should be understood that changes, additions, and omissions to that sequence can occur without substantially affecting the operation of the disclosed configuration and aspects.

[0091] Several variations and modifications of this disclosure may be used. Some features of this disclosure may be provided without providing others.

[0092] In another configuration, the systems and methods of this disclosure may be implemented in combination with a dedicated computer, a programmable microprocessor or microcontroller and (multiple) peripheral integrated circuit elements, an ASIC or other integrated circuit, a digital signal processor, hardwired electronic or logic circuits such as discrete component circuits, programmable logic devices or gate arrays such as PLDs, PLAs, FPGAs, PALs, dedicated computers, any similar components, etc. Generally, any (multiple) devices or components capable of implementing the methods described herein can be used to implement various aspects of this disclosure. Example hardware that can be used in this disclosure includes computers, handheld devices, telephones (e.g., cellular, internet-enabled, digital, analog, hybrid, etc.), and other hardware known in the art. Some of these devices include processors (e.g., single-processor or multi-microprocessor), memory, non-volatile storage devices, input devices, and output devices. Furthermore, alternative software implementations, including but not limited to distributed processing or component / object distributed processing, parallel processing, or virtual machine processing, can be constructed to implement the methods described herein.

[0093] In another configuration, the disclosed method can be readily implemented using software from an object-oriented or object-centered software development environment that provides portable source code usable on various computer or workstation platforms. Alternatively, the disclosed system can be partially or fully implemented in hardware using standard logic circuitry or VLSI design. Whether software or hardware is used to implement a system according to this disclosure depends on the system's speed and / or efficiency requirements, specific functionality, and the particular software or hardware system or microprocessor or microcomputer system utilized.

[0094] In another configuration, the disclosed methods can be implemented in part in software, which can be stored on a storage medium and executed on a programmed general-purpose computer using the cooperation of a controller and memory, a special-purpose computer, a microprocessor, etc. In these cases, the systems and methods of this disclosure can be implemented as programs embedded in a personal computer (such as applets, JAVA®, or CGI scripts), resources residing on a server or computer workstation, routines embedded in a dedicated measurement system, system components, etc. The system can also be implemented by physically incorporating the system and / or methods into a software system and / or hardware system.

[0095] This disclosure is not limited to the standards and protocols described. Other similar standards and protocols not mentioned herein exist and are included in this invention. Furthermore, the standards and protocols mentioned herein, as well as other similar standards and protocols not mentioned herein, are periodically replaced by equivalent standards and protocols that are substantially the same in function but faster or more efficient. Such alternative standards and protocols with the same functionality are considered equivalents included in this disclosure.

[0096] According to at least one example of this disclosure, a method for guiding the behavior of a generative artificial intelligence (AI) system is provided. The method may include: obtaining an input cue word associated with a requested task for one or more generative AI systems; obtaining one or more attributes based on the input cue word; modifying the input cue word based on the one or more embedded attributes; and providing the modified input cue word to one or more generative AI systems.

[0097] According to at least one aspect of the above method, the method may include obtaining one or more embedding attributes based on an input prompt word, including determining whether the one or more embedding attributes are embedded in one or more applications, documents, interfaces, and / or content, which are intended to communicate with one or more generative AI systems to perform a requested task based on an input prompt word.

[0098] According to at least one aspect of the above method, the method may include one or more embedded attributes being received via one or more development tools for generating documents or other types of content and / or applications or interfaces.

[0099] According to at least one aspect of the above method, the method may include one or more embedded attributes including one or more tags, prompts, markers or instructions that provide additional instructions to one or more generative AI systems.

[0100] According to at least one aspect of the above method, the method may include one or more of the embedded attributes being embedded in an online forum or social media site and designed to guide the tone of discussion or guide writing aids to follow the rules of the online forum or social media site.

[0101] According to at least one aspect of the above method, the method may include one or more of the embedded attributes being different embedded attributes around each text field on the webpage, and may include examples that can be used as appropriate responses to guidance for one or more generative AI systems.

[0102] According to at least one aspect of the above method, the method may include one or more generative AI systems comprising one or more generative large language machine learning models, one or more transformer models, and / or combinations of machine learning models.

[0103] According to at least one example of this disclosure, a method for guiding the behavior of a generative artificial intelligence (AI) system is provided. The method may include: obtaining input prompts associated with a requested task for one or more generative AI systems; obtaining one or more attributes based on the input prompts; generating supplementary prompts based on the one or more attributes in response to determining the existence of the attributes; and providing the supplementary prompts and the input prompts to the one or more generative AI systems.

[0104] According to at least one aspect of the above method, the method may include obtaining one or more embedding attributes based on an input prompt word, including determining whether the one or more embedding attributes are embedded in one or more applications, documents, interfaces, and / or content, which are intended to communicate with one or more generative AI systems to perform a requested task based on an input prompt word.

[0105] According to at least one aspect of the above method, the method may include one or more embedded attributes being received via one or more development tools for generating documents or other types of content and / or applications or interfaces.

[0106] According to at least one aspect of the above method, the method may include one or more embedding attributes embedded around an HTML text area of ​​the email body, which provides guidance to the AI ​​to compose in the same tone as other emails previously sent to a particular recipient.

[0107] According to at least one aspect of the above method, the method may include one or more of the embedded attributes being embedded in an online forum or social media site and designed to guide the tone of discussion or guide writing aids to follow the rules of the online forum or social media site.

[0108] According to at least one aspect of the above method, the method may include one or more of the embedded attributes being different embedded attributes around each text field on the webpage, and may include examples that can be used as appropriate responses to guidance for one or more generative AI systems.

[0109] According to at least one example of this disclosure, a computing device is provided for guiding the behavior of a generative artificial intelligence (AI) system. The computing device may include a processor and a memory having a plurality of instructions stored thereon, which, when executed by the processor, cause the computing device to: obtain, for one or more generative AI systems, an input cue word associated with a requested task; obtain one or more attributes based on the input cue word; modify the input cue word based on the one or more embedded attributes; and provide the modified input cue word to the one or more generative AI systems.

[0110] According to at least one aspect of the computing device described above, the computing device may include obtaining one or more embedded attributes based on an input prompt word, including determining whether the one or more embedded attributes are embedded in one or more applications, documents, interfaces, and / or content, which are intended to communicate with one or more generative AI systems to perform a requested task based on the input prompt word.

[0111] According to at least one aspect of the computing device described above, the computing device may include one or more embedded attributes that are received via one or more development tools for generating documents or other types of content and / or applications or interfaces.

[0112] According to at least one aspect of the computing device described above, the computing device may include one or more embedded attributes embedded around an HTML text area of ​​an email body that provides guidance to the AI ​​to compose in the same tone as other emails previously sent to a particular recipient.

[0113] According to at least one aspect of the computing device described above, the computing device may include one or more embedded attributes that are embedded in an online forum or social media site and are designed to guide the tone of discussion or guide writing assistance to follow the rules of the online forum or social media site.

[0114] According to at least one aspect of the computing device described above, the computing device may include one or more embedded attributes that are different embedded attributes around each text field on a webpage, and may include examples that can be used as appropriate responses to guidance for one or more generative AI systems.

[0115] According to at least one aspect of the computing device described above, the computing device may include one or more generative AI systems comprising one or more generative large language machine learning models, one or more transformer models, and / or combinations of machine learning models.

[0116] In various configurations and aspects, this disclosure includes components, methods, processes, systems, and / or apparatuses substantially as depicted and described herein, including various combinations, sub-combinations, and subsets thereof. Upon understanding this disclosure, those skilled in the art will understand how to make and use the systems and methods disclosed herein. In various configurations and aspects, apparatuses and processes are provided in the absence of items not depicted and / or described herein and / or in various configurations or aspects thereof (including in the absence of such items that may have been used in prior apparatus or processes, for example, to improve performance, ease of implementation, and / or reduce implementation costs).

Claims

1. A method for guiding the behavior of a generative artificial intelligence (AI) system (160), the method comprising: For one or more generative AI systems (160), obtain input prompts associated with the requested task; One or more attributes are obtained based on the input prompt words; Modify the input prompt word based on the one or more embedded attributes; as well as The modified input prompts are provided to the one or more generative AI systems (160).

2. The method of claim 1, wherein obtaining one or more embedding attributes based on the input prompt word includes determining whether the one or more embedding attributes are embedded in one or more applications, documents, interfaces and / or content, the one or more applications, documents, interfaces and / or content being designed to communicate with the one or more generative AI systems (160) to perform the requested task based on the input prompt word.

3. The method of claim 2, wherein the one or more embedded attributes are received via one or more development tools for generating documents or other types of content and / or applications or interfaces.

4. The method of claim 1, wherein the one or more embedded attributes include one or more tags, prompts, markers or indications that provide additional instructions to the one or more generative AI systems (160).

5. The method of claim 1, wherein the one or more embedded attributes are embedded in an online forum or social media site and are designed to guide the tone of discussion or guide writing to comply with the rules of the online forum or social media site.

6. The method of claim 1, wherein the one or more embedded attributes are different embedded attributes around each text field on the webpage, and include examples that can be used as appropriate responses to guidance of the one or more generative AI systems (160).

7. The method of claim 1, wherein the one or more generative AI systems (160) comprises one or more generative large language machine learning models, one or more transformer models and / or a combination of machine learning models.

8. A method for guiding the behavior of a generative artificial intelligence (AI) system (160), the method comprising: For one or more generative AI systems (160), obtain input prompts associated with the requested task; One or more attributes are obtained based on the input prompt words; In response to determining that the one or more attributes exist, supplementary prompt words are generated based on the one or more attributes; as well as The supplementary prompts and the input prompts are provided to the one or more generative AI systems (160).

9. The method of claim 8, wherein obtaining one or more embedding attributes based on the input prompt word includes determining whether the one or more embedding attributes are embedded in one or more applications, documents, interfaces and / or content, the one or more applications, documents, interfaces and / or content being designed to communicate with the one or more generative AI systems (160) to perform the requested task based on the input prompt word.

10. The method of claim 9, wherein the one or more embedded attributes are received via one or more development tools for generating documents or other types of content and / or applications or interfaces.

11. The method of claim 8, wherein one or more embedding attributes are embedded around an HTML text area of ​​the email body, the email body providing guidance to the AI ​​to compose in the same tone as other emails previously sent to a particular recipient.

12. The method of claim 8, wherein the one or more embedded attributes are embedded in an online forum or social media site and are designed to guide the tone of discussion or guide writing to follow the rules of the online forum or social media site.

13. The method of claim 8, wherein the one or more embedded attributes are different embedded attributes around each text field on the webpage, and include examples that can be used as appropriate responses to guidance of the one or more generative AI systems (160).

14. A computing device (130) for guiding the behavior of a generative artificial intelligence (AI) system (160), said computing device comprising: processor; as well as A memory having a plurality of instructions stored thereon, which, when executed by the processor, cause the computing device to: For one or more generative AI systems (160), obtain input prompts associated with the requested task; One or more attributes are obtained based on the input prompt words; Modify the input prompt word based on the one or more embedded attributes; as well as The modified input prompts are provided to the one or more generative AI systems (160).

15. The computing device (130) of claim 14, wherein obtaining one or more embedding attributes based on the input prompt word includes determining whether the one or more embedding attributes are embedded in one or more applications, documents, interfaces and / or content, the one or more applications, documents, interfaces and / or content being designed to communicate with the one or more generative AI systems (160) to perform the requested task based on the input prompt word.

16. The computing device (130) of claim 15, wherein the one or more embedded attributes are received via one or more development tools for generating documents or other types of content and / or applications or interfaces.

17. The computing device (130) of claim 14, wherein one or more embedding attributes are embedded around an HTML text area of ​​an email body that provides guidance to the AI ​​to compose in the same tone as other emails previously sent to a particular recipient.

18. The computing device (130) of claim 14, wherein one or more embedded attributes are embedded in an online forum or social media site and are designed to guide the tone of discussion or guide writing aids to follow the rules of the online forum or social media site.

19. The computing device (130) of claim 14, wherein the one or more embedding attributes are different embedding attributes around each text field on the webpage, and include examples that can be used as appropriate responses to guidance of the one or more generative AI systems (160).

20. The computing device (130) of claim 14, wherein the one or more generative AI systems (160) comprise one or more generative large language machine learning models, one or more transformer models and / or combinations of machine learning models.

Images