Agent-based creative scheme generation method, apparatus, device, and medium

Abstract

The present disclosure discloses an agent-based creative scheme generation method and device, equipment and medium, which relates to the technical field of artificial intelligence such as agent and natural language processing. The specific implementation scheme is: based on the input information of the user, obtaining reference information and creative target for generating creative scheme; the reference information includes at least one of specified paper and specified topic; obtaining a set of verification information for verifying the creative scheme; using an agent, based on the reference information, the creative target and the set of verification information, generating a plurality of first candidate creative schemes; based on the set of verification information, screening a first target creative scheme from the plurality of first candidate creative schemes.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, specifically to the field of artificial intelligence technology such as intelligent agents and natural language processing, and particularly to a method, apparatus, device and medium for generating creative solutions based on intelligent agents. Background Technology

[0002] In existing technologies, intelligent agents can be applied to personal office work, customer service systems in various fields, and various analysis and prediction systems. Moreover, with the development of technology, intelligent agent technology is becoming increasingly mature, enabling various industries to improve the efficiency of information processing. Summary of the Invention

[0003] This disclosure provides a method, apparatus, device, and medium for generating creative solutions based on intelligent agents.

[0004] According to one aspect of this disclosure, a method for generating creative solutions based on intelligent agents is provided, comprising:

[0005] Based on user input, reference information and creative goals for generating creative solutions are obtained; the reference information includes at least one of a specified paper and a specified topic.

[0006] Obtain a set of verification information for the creative solution;

[0007] Using an intelligent agent, multiple first candidate creative solutions are generated based on the reference information, the creative goal, and the set of verification information;

[0008] Based on the set of verification information, a first target creative solution is selected from the plurality of first candidate creative solutions.

[0009] According to another aspect of this disclosure, an agent-based creative solution generation apparatus is provided, comprising:

[0010] The input acquisition module is used to acquire reference information and creative goals for generating creative solutions based on user input information; the reference information includes at least one of a specified paper and a specified topic;

[0011] The collection acquisition module is used to acquire a collection of verification information for validating creative solutions;

[0012] The generation module is used to generate multiple first candidate creative solutions using an intelligent agent based on the reference information, the creative target, and the set of verification information.

[0013] The filtering module is used to filter the first target creative scheme from the plurality of first candidate creative schemes based on the set of verification information.

[0014] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0015] At least one processor; and

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods described above and any possible implementations.

[0018] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing the computer to perform the methods described above and any possible implementation thereof.

[0019] According to another aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the aspects and any possible implementations described above.

[0020] According to the technology disclosed herein, creative solutions can be generated accurately and efficiently.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0022] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0023] Figure 1 This is a schematic diagram based on the first embodiment of the present disclosure;

[0024] Figure 2 This is a schematic diagram according to the second embodiment of the present disclosure;

[0025] Figure 3 This is a schematic diagram according to the third embodiment of the present disclosure;

[0026] Figure 4 This is a schematic diagram according to the fourth embodiment of the present disclosure;

[0027] Figure 5 This is a block diagram of an electronic device used to implement the methods of the embodiments of this disclosure. Detailed Implementation

[0028] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0029] Obviously, the described embodiments are only some, not all, of the embodiments disclosed herein. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure.

[0030] It should be noted that the terminal devices involved in the embodiments of this disclosure may include, but are not limited to, smart devices such as mobile phones, personal digital assistants (PDAs), wireless handheld devices, and tablet computers; the display devices may include, but are not limited to, personal computers, televisions, and other devices with display functions.

[0031] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0032] Figure 1 This is a schematic diagram based on the first embodiment of the present disclosure; as shown Figure 1 As shown, this embodiment provides a method for generating creative solutions based on intelligent agents, which may specifically include the following steps:

[0033] S101. Based on the user's input information, obtain reference information and creative goals for generating creative solutions; the reference information includes at least one of a specified paper and a specified topic;

[0034] The technical solution of this embodiment is used to assist users in generating efficient creative solutions. In specific applications, users need to first input information for generating creative solutions. For example, this input information may implicitly or explicitly include creative goals, and may also include reference information. For example, in the scenario of this embodiment, the reference information may be a specified paper, which can be any publicly available literature. In practical applications, the specified paper may include a clear research field, research topic, and detailed solution information. By inputting the specified paper, the intelligent agent can generate a target creative solution that meets the creative goals based on the specified paper.

[0035] Alternatively, the reference information may include a specified topic, which may include a clear research field, research direction, and a brief overview of the research plan. By inputting this specified topic, the agent can also generate a target creative solution that meets the creative objectives.

[0036] In practical applications, the reference information may include only one of the specified paper and the specified topic, or it may include both. This provides the agent with more detailed reference information, enabling the agent to generate more accurate and effective creative solutions that meet the creative objectives.

[0037] S102. Obtain the set of verification information for the creative solution;

[0038] In this embodiment, to improve the accuracy of the final generated target creative solution, it is also necessary to verify the initially generated creative solution. Therefore, it is also necessary to obtain a set of verification information for verifying the creative solution.

[0039] In practical applications, the methods for verifying solutions may differ across different research directions in different fields. Therefore, in this embodiment, the set of verification information may vary depending on the specific papers and / or topics specified in the reference information.

[0040] S103. Using an intelligent agent, multiple first candidate creative solutions are generated based on a set of reference information, creative goals, and verification information.

[0041] For example, in a practical implementation, an intelligent agent can generate multiple first-candidate creative solutions that can be verified through a set of verification information based on reference information and creative goals.

[0042] S104. Based on the set of verification information, select the first target creative solution from multiple first candidate creative solutions.

[0043] Finally, when selecting the first target creative solution from multiple first candidate creative solutions, the selection is again based on the set of verification information to improve the accuracy of the final first target creative solution.

[0044] The executing entity of the agent-based creative solution generation method in this embodiment can be an agent-based creative solution generation device. This device can be an electronic entity or an application integrated with software. In use, the user can input information into the device, which can ultimately generate and output a first target creative solution for the user's reference.

[0045] The technical solution of this embodiment can be applied to scientific research support platforms, technical intelligence analysis platforms, and enterprise internal R&D decision-making platforms. It is especially suitable for scenarios where subsequent innovation schemes are deduced, schemes are screened, and research routes are planned around specific papers, specific research topics, or specific technical problems, and can effectively improve the accuracy of the generated creative schemes.

[0046] The agent-based creative solution generation method of this embodiment can accurately and effectively generate multiple first candidate creative solutions by employing an agent based on a set of reference information, creative goals, and verification information. Furthermore, it can filter a first target creative solution from these multiple first candidate solutions based on the verification information set, further improving the accuracy of the filtering and thus the accuracy of the obtained first target innovative solution. Therefore, the technical solution of this embodiment can generate creative solutions accurately and efficiently.

[0047] Figure 2 This is a schematic diagram according to the second embodiment of the present disclosure; the agent-based creative solution generation method of this embodiment, in the above... Figure 1 Based on the technical solutions of the illustrated embodiments, the technical solutions of this disclosure can be described in more detail. For example... Figure 2 As shown, the agent-based creative solution generation method of this embodiment may specifically include the following steps:

[0048] S201. Based on the user's input information, obtain reference information and creative goals for generating creative solutions; the reference information includes at least one of a specified paper and a specified topic;

[0049] For example, based on user input, an intelligent agent can extract reference information from the input to generate creative solutions. Simultaneously, the agent can also extract creative goals from the input. For instance, the user's input could be: "Please refer to paper 1 and help generate some creative solutions to solve problem B." Or, the user's input could be: "Help generate some researchable creative ideas about topic E in domain D." Alternatively, it could be: "Please refer to paper 2 and help generate some creative ideas about topic F," and so on.

[0050] It should be noted that, in the specific implementation of this step in this embodiment, reference information and creative goals can be obtained through a single interaction with the user on the application interface of the intelligent agent-based creative solution generation device, or reference information and creative goals can be obtained through multiple interactions with the user on the application interface.

[0051] S202. Employ an intelligent agent to generate a set of reference solutions based on input information;

[0052] S203. Using an intelligent agent, obtain a set of verification datasets and a set of verification metrics based on a set of reference schemes.

[0053] In this embodiment, taking the verification information set including a reference scheme set, a verification dataset set, and a verification indicator set as an example, the verification information set constructed in this way includes comprehensive and rich content.

[0054] The reference scheme set may include multiple reference schemes related to the reference information. If the reference information includes a specific paper, the reference scheme set may include that specific paper. The validation dataset set includes multiple validation datasets used to validate the reference schemes and creative schemes. For example, by analyzing each reference scheme in the reference scheme set, validation datasets in the relevant field can be obtained. Alternatively, by analyzing each reference scheme and the reference information, the relevant field can be obtained, and then multiple validation datasets in the relevant field can be obtained from the network. The validation metric set may include validation metrics used when validating the reference schemes and creative schemes using datasets from the validation dataset set. Specifically, the validation metric set may include multiple validation metrics, one of which is the primary validation metric.

[0055] In practical applications, in some specialized fields, the validation information set can take other forms. For example, it may only include the validation dataset set and the main validation metric. Alternatively, it may include multiple validation tools, etc., without limitation. To improve validation efficiency, the minimum number of reference schemes included in the reference scheme set and the minimum number of validation information sets included in the validation information set can be limited.

[0056] In practical applications, if the user's input information is rich enough, including all the information in the verification information set, then the verification information set can be obtained directly from the user's input information.

[0057] If the user's input information is insufficient to directly obtain a set of verification information, an intelligent agent can be used to generate a set of reference solutions based on the input information. For example, in a practical implementation, an intelligent agent can retrieve multiple relevant reference solutions from a specified database based on a given paper and creative goal; then, based on the specified paper and multiple relevant reference solutions, a set of reference solutions can be constructed, effectively enriching the set and improving its accuracy. The specified database can be a database in the corresponding field determined based on the specified paper and creative goal.

[0058] Alternatively, in practical applications, users can also input a small number of initial reference schemes in the input information, but the number should not reach the minimum number in the reference scheme set. For example, the specified paper in the reference information in the above embodiment is an initial reference scheme. In practical applications, the input information may include other initial reference schemes in addition to the specified paper. In this case, the initial reference schemes can be obtained from the input information first, and then an intelligent agent can retrieve multiple relevant reference schemes from a specified database based on the initial reference schemes, the specified paper, and the creative goal. Then, based on the multiple relevant reference schemes and the initial reference schemes, a reference scheme set can be constructed to effectively enrich the reference schemes in the reference scheme set and improve the comprehensiveness and accuracy of the reference scheme set.

[0059] S204. An intelligent agent is used to generate gap information based on reference information, each reference scheme in the reference scheme set, and the creative goal; the gap information is used to indicate the problems that exist in achieving the creative goal.

[0060] For example, an intelligent agent can intelligently analyze various reference solutions based on creative goals and reference information, determine the problems existing in each reference solution when achieving the creative goals, and obtain the gap information corresponding to each reference solution.

[0061] S205. Using an intelligent agent, multiple first candidate creative solutions are generated based on the creative goal, reference information, gap information, and each reference solution in the reference solution set.

[0062] Specifically, an intelligent agent can be used, guided by gap information and with creative goals as its objective, to generate creative solutions based on reference information and various reference solutions, thereby obtaining multiple first candidate creative solutions.

[0063] Alternatively, this step may further include the following steps in its implementation:

[0064] (1) Using an intelligent agent, multiple initial creative solutions are generated based on the creative goal, reference information, gap information and each reference solution in the reference solution set;

[0065] (2) Using an intelligent agent, multiple first extended creative solutions are generated based on multiple initial creative solutions;

[0066] (3) Based on multiple initial creative schemes and multiple first extended creative schemes, generate multiple first candidate creative schemes.

[0067] In other words, in practical applications, the multiple initial creative schemes generated in step (1) can be directly used as multiple first candidate creative schemes. However, the number of first candidate creative schemes obtained at this time may not be rich enough, and the creativity may not be comprehensive enough. Furthermore, an agent can be used to extend the multiple initial creative schemes. For example, during the extension, different initial creative schemes can be combined with each other to generate new first extended creative schemes, thereby obtaining multiple first extended creative schemes.

[0068] This implementation method can generate multiple first-choice creative solutions more accurately and comprehensively.

[0069] Further optionally, in one embodiment of this disclosure, in step (3), multiple initial creative solutions and multiple first extended creative solutions can be directly combined as multiple first candidate creative solutions.

[0070] Further, optionally, in order to improve the accuracy of the generated multiple first candidate creative solutions, step (3) can also be implemented using the following steps:

[0071] (a1) Construct a set of alternative creative solutions by taking multiple initial creative solutions and multiple first extended creative solutions as alternative creative solutions;

[0072] (b1) Based on the set of verification information, each candidate creative solution is comprehensively scored to obtain the first score of each candidate creative solution;

[0073] (c1) Based on the first score of each alternative creative solution, obtain multiple alternative creative solutions from the set of alternative creative solutions whose first score is greater than the first preset threshold, and use them as multiple first candidate creative solutions.

[0074] In this implementation, multiple initial creative solutions and multiple first extended creative solutions can be combined to form a set of candidate creative solutions. Then, a validation dataset is used to validate each candidate creative solution, thereby selecting multiple first candidate creative solutions from the set of candidate creative solutions.

[0075] For example, in the specific implementation of step (b1), each reference scheme in the reference scheme set of the verification information set can be used as a reference benchmark. Based on the partial data in each verification dataset and the verification index set in the verification dataset set, each candidate creative scheme can be comprehensively scored to obtain the first score of each candidate creative scheme.

[0076] Specifically, this step is implemented in the stage of screening multiple first candidate creative solutions from the set of alternative creative solutions. At this time, in order to improve the verification efficiency, when using each verification dataset in the verification dataset set to comprehensively score each alternative creative solution, only a portion of the data in each verification dataset can be used. For example, according to actual needs or experience, a portion of the data in each verification dataset can be selected in proportions such as 1 / 10, 1 / 8, or 1 / 5.

[0077] For example, in one embodiment of this disclosure, step (b) may include the following steps:

[0078] (a2) Using partial data from each verification dataset, verify the main verification indicators in the verification indicator set corresponding to each alternative creative scheme and each reference scheme, and obtain the values ​​of the main verification indicators obtained when each verification dataset verifies each alternative creative scheme and each reference scheme, as well as the verification runtime when each alternative creative scheme and each reference scheme is verified.

[0079] For each alternative creative solution or each reference solution, the corresponding verification runtime can be the total runtime of verifying the solution using all verification datasets in the verification dataset set; or it can be the average runtime of verifying the solution using all verification datasets in the verification dataset set.

[0080] Specifically, when verifying each alternative creative solution, the alternative creative solution is compared with multiple reference solutions in the reference solution set. Since all alternative creative solutions are compared with the same multiple reference solutions, that is, the comparison and verification standards are the same, the verification results can be accurately and effectively identified, and the first score of each alternative creative solution can be obtained more accurately.

[0081] To improve verification efficiency, since the reference schemes, verification datasets, and master verification metrics are all known, the values ​​of the master verification metrics corresponding to each reference scheme can be verified in advance using the respective verification datasets. Simultaneously, the verification runtime of each reference scheme should also be calculated.

[0082] When validating each alternative creative solution, we use each validation dataset to verify the value of the main validation metric corresponding to each alternative solution and obtain the validation runtime of each alternative creative solution.

[0083] (b2) Based on the values ​​of the main validation metrics obtained when validating each candidate creative solution and each reference solution using each validation dataset, obtain the ranking score of each candidate creative solution;

[0084] Specifically, for each candidate creative solution, the ranking of the candidate creative solution on each validation dataset is determined based on the values ​​of the main validation metrics obtained when validating the candidate creative solution and each reference solution on each validation dataset; for each candidate creative solution, the average of the inverses of the rankings when validating on each validation dataset is taken as the ranking score of the corresponding candidate creative solution.

[0085] For example, given a given alternative creative solution, if the validation dataset includes three datasets, the values ​​of the primary validation metrics for the alternative creative solution validated in validation dataset 1, and the values ​​of the primary validation metrics for each reference solution in validation dataset 1, are sorted in ascending order to determine the alternative creative solution's ranking as number 1. Similarly, the values ​​of the primary validation metrics for the alternative creative solution validated in validation dataset 2, and the values ​​of the primary validation metrics for each reference solution in validation dataset 2, are sorted in ascending order to determine the alternative creative solution's ranking as number 1. Finally, the values ​​of the primary validation metrics for the alternative creative solution validated in validation dataset 3, and the values ​​of the primary validation metrics for each reference solution in validation dataset 3, are sorted in ascending order to determine the alternative creative solution's ranking as number 3. The ranking score of this alternative creative solution can be equal to 1 / 3 (1 + 1 + 1 / 3).

[0086] It should be noted that this embodiment uses the calculation of only the main verification indicator in the verification indicator set during verification. In practical applications, more verification indicators in the verification indicator set can also be calculated during verification. The calculation method for each verification indicator is the same.

[0087] Alternatively, the values ​​of multiple verification indicators can be weighted and summed to obtain the comprehensive verification indicator value for each reference scheme and the comprehensive verification indicator value for each candidate creative scheme. Then, the ranking score for each candidate creative scheme is calculated according to the method described in the above embodiment, which will not be repeated here.

[0088] Alternatively, other verification metrics can be used as criteria for selecting multiple first-candidate creative solutions. For example, a candidate creative solution can be selected as a first-candidate creative solution only if its first score is greater than a first preset threshold and the value of a second verification metric (in addition to the main verification metric) also meets the preset conditions. Alternatively, when multiple first-candidate creative solutions are selected and presented to the user, the values ​​of all verification metrics for each first-candidate creative solution can be displayed simultaneously. This allows the user to manually check whether the values ​​of each verification metric meet the requirements and further eliminate first-candidate creative solutions that do not meet the requirements based on the values ​​of each verification metric, thereby improving the accuracy of the final selection of multiple first-candidate creative solutions.

[0089] Using the above method, the ranking scores of each alternative creative solution can be obtained accurately and efficiently.

[0090] (c2) Based on the verification runtime corresponding to each alternative creative solution and each reference solution, obtain the penalty score for each alternative creative solution;

[0091] For example, in the specific implementation of this step, the maximum and minimum verification runtimes among the verification runtimes of each candidate creative solution can be obtained first; then, based on the verification runtimes of each candidate creative solution, as well as the maximum and minimum verification runtimes, the penalty score for each candidate creative solution can be determined using the following formula:

[0092] Penalty(x) = max(0, (Time(x) - Time_min_base) / (Time_max_base -Time_min_base));

[0093] x is the alternative creative solution, Penalty(x) is the penalty score of alternative creative solution x, Time_min_base is the minimum verification runtime among multiple reference placement verification runtimes, Time_max_base is the maximum verification runtime among multiple reference placement verification runtimes; Time(x) is the verification runtime of alternative creative solution x.

[0094] By adopting this penalty score setting method, if the verification runtime of a candidate creative solution is greater than the minimum verification runtime, it indicates that the candidate creative solution still has room for improvement. In this case, the difference between the verification runtime of the candidate creative solution and the minimum verification runtime can be divided by the ratio of the difference between the maximum verification runtime and the minimum verification runtime. That is, the penalty score of the candidate creative solution is configured according to an appropriate ratio. Conversely, if the verification runtime of a candidate creative solution is less than or equal to the minimum verification runtime, it indicates that the verification of the candidate creative solution is already very good, and the corresponding penalty score can be 0.

[0095] By adopting the above method, the penalty scores for each alternative creative solution can be set accurately and reasonably.

[0096] (d2) For each alternative creative solution, the ranking score and the penalty score of the alternative creative solution are weighted and subtracted to obtain the first score of the alternative creative solution.

[0097] Specifically, the weights of ranking scores and penalty scores can be configured based on experience or needs. For example, if the weight of the ranking score is a1 and the weight of the penalty score is a2, the first score of the alternative creative solution = ranking score * a1 - penalty score * a2.

[0098] Finally, multiple candidate creative solutions with a first score greater than a first preset threshold are selected from the set of alternative creative solutions and used as multiple first candidate creative solutions. The first preset threshold can be set based on experience or actual needs and is not limited here.

[0099] S206. Based on the verification information set, a comprehensive score is given to each first candidate creative solution to obtain the second score value of each first candidate creative solution.

[0100] Specifically, when implementing this step, each reference scheme in the reference scheme set can be used as a reference benchmark. Based on the full data and verification index set in each verification dataset in the verification dataset set, each first candidate creative scheme is comprehensively scored to obtain the second score of each first candidate creative scheme.

[0101] In this embodiment, the specific implementation of this step is similar to that of step (b1) described above, except that step (b1) is located during the process of screening multiple first candidate creative schemes, and is not the final target creative scheme at this time. To improve screening efficiency, only a portion of the data can be selected when using each verification dataset for verification. However, step S206 is located during the process of screening the first target creative scheme. To improve the accuracy of screening, all data is selected when using each verification dataset for verification. The remaining implementation methods are the same as those in the above embodiments, and details can also be found in the relevant descriptions of the above embodiments.

[0102] For example, in practical implementation, the following steps can be taken:

[0103] (a3) Using the full data in each verification dataset, verify the main verification indicators in the verification indicator set corresponding to each first candidate creative scheme and each reference scheme, and obtain the values ​​of the main verification indicators obtained when verifying each first candidate creative scheme and each reference scheme in each verification dataset, as well as the verification runtime when verifying each first candidate creative scheme and each reference scheme.

[0104] (b3) Based on the values ​​of the main verification metrics obtained when verifying each first candidate creative scheme and each reference scheme using each verification dataset, obtain the ranking score of each first candidate creative scheme;

[0105] For example, in the specific implementation of step (b3), for each first candidate creative scheme, the ranking of the first candidate creative scheme when using each verification dataset can be determined based on the values ​​of the main verification indicators obtained when verifying the first candidate creative scheme and each reference scheme on each verification dataset; for each first candidate creative scheme, the average of the inverses of the rankings of the first candidate creative schemes when verifying on each verification dataset is taken as the ranking score of the corresponding first candidate creative scheme.

[0106] (c3) Based on the verification runtime corresponding to each first candidate creative solution and each reference solution, obtain the penalty score of each first candidate creative solution;

[0107] For example, we can first obtain the maximum and minimum verification runtimes among the verification runtimes of each reference solution; then, based on the verification runtimes of each first candidate creative solution, as well as the maximum and minimum verification runtimes, we can determine the penalty score for each first candidate creative solution using the following formula:

[0108] Penalty(h) = max(0, (Time(h) - Time_min_base) / (Time_max_base -Time_min_base));

[0109] h is the first candidate creative solution, Penalty(h) is the penalty score of the first candidate creative solution h, Time_min_base is the minimum verification runtime, Time_max_base is the maximum verification runtime, and Time(h) is the verification runtime of the first candidate creative solution h.

[0110] (d3) For each first candidate creative solution, the ranking score and the penalty score of the first candidate creative solution are weighted and subtracted to obtain the second score of the first candidate creative solution.

[0111] In this embodiment, the specific implementation principle of steps (a3)-(d3) is the same as that of steps (a2)-(d2) above. For details, please refer to the relevant records of the above embodiments, which will not be repeated here.

[0112] By using the above method, the second score of each first candidate creative solution can be obtained accurately and efficiently.

[0113] S207. Based on the second score of each first candidate creative solution and the second preset threshold, obtain the first target creative solution from multiple first candidate creative solutions.

[0114] The second preset threshold can be set according to actual needs or experience. In this embodiment, the number of first target creative solutions obtained can be one, two, or more.

[0115] Further optionally, after step S207, the method may include: showing the user the final obtained first target creative solution so that the user can carry out research and development based on the first target creative solution.

[0116] Optionally, in one embodiment of this disclosure, the values ​​of each verification indicator corresponding to the first target creative solution can also be displayed to the user simultaneously when verifying the first target creative solution using each verification dataset, so that the user can understand the verification information of the first target creative solution.

[0117] According to the above scheme of this embodiment, the first target creative scheme can be obtained accurately and efficiently through the verification of the verification information set.

[0118] S208. Employ an intelligent agent to periodically detect whether the reference scheme set needs to be updated;

[0119] S209. In response to the need to update the reference scheme set, obtain the updated reference scheme; and add the updated reference scheme to the reference scheme set.

[0120] Specifically, an intelligent agent can be used to periodically detect whether reference information and / or the latest solutions related to each reference solution appear in the network. If so, it is confirmed that the reference solution set needs to be updated. At this time, the latest solution is obtained as the updated reference solution, and then the updated reference solution is added to the reference solution set, which can improve the accuracy of the reference solution set. In this embodiment, the number of updated reference solutions is not limited.

[0121] S210. Using an intelligent agent, based on the updated reference scheme and the first target creative scheme, generate multiple second extended creative schemes;

[0122] Specifically, the intelligent agent can analyze the updated reference scheme and the first target creative scheme to discover some creative schemes that can be combined or derived, and then generate multiple second extended creative schemes.

[0123] S211. Based on the first target creative solution and multiple second extended creative solutions, generate multiple second candidate creative solutions;

[0124] When implementing this step, you can refer to the specific implementation of step (3) in the above embodiment, which will not be repeated here.

[0125] S212. Based on the full data and verification index set in each verification dataset in the verification dataset set, a comprehensive score is given to each second candidate creative scheme to obtain the third score value of each second candidate creative scheme.

[0126] When implementing this step, you can refer to the specific implementation of step 206 in the above embodiment, which will not be repeated here.

[0127] S213. Based on the third score of each second candidate creative scheme and the second preset threshold, obtain the second target creative scheme from the first target creative scheme and multiple second extended creative schemes.

[0128] When implementing this step, you can refer to the specific implementation of step 207 in the above embodiment, which will not be repeated here.

[0129] Based on the above, it can be seen that this embodiment can continuously detect after generating the first target creative solution, and update and generate the second target creative solution using the above steps S208-S213 when updates are needed. Similarly, after step S213, steps S208-S213 can be referred to to continuously detect and update, which will not be elaborated here.

[0130] Optionally, in this embodiment, the target creative solutions generated each time, such as the first target creative solution and the second target creative solution, can be retained, and information such as which updated reference files were introduced as the reason for each update can be recorded so that they can be traced later.

[0131] This approach allows for timely updates to the generated target creative solutions, effectively improving the accuracy of the final target creative solutions and mitigating the risk of unreasonable creative solutions consuming R&D resources.

[0132] The agent-based creative solution generation method of this embodiment can provide accurate, effective, and valuable creative solutions in scientific research assistance platforms, enterprise R&D platforms, and automated algorithm discovery platforms, effectively promoting R&D progress and improving R&D efficiency.

[0133] The agent-based creative solution generation method in this embodiment can significantly reduce the risk of "rashly generating ideas based on only a few similar papers". It can use agents to generate and verify reasonable and effective creative solutions based on a set of verification information, thereby improving the accuracy and effectiveness of the generated creative solutions.

[0134] Figure 3 This is a schematic diagram based on the third embodiment of this disclosure; as shown Figure 3 As shown, this embodiment provides a creative solution generation device 300 based on an intelligent agent, including:

[0135] The input acquisition module 301 is used to acquire reference information and creative goals for generating creative solutions based on user input information; the reference information includes at least one of a specified paper and a specified topic;

[0136] The collection acquisition module 302 is used to acquire the collection of verification information for the creative scheme.

[0137] The generation module 303 is used to generate multiple first candidate creative solutions using an intelligent agent based on the reference information, the creative target, and the set of verification information.

[0138] The filtering module 304 is used to filter the first target creative scheme from the plurality of first candidate creative schemes based on the verification information set.

[0139] The agent-based creative idea generation device 300 of this embodiment achieves the same implementation principle and technical effect as the agent-based creative idea generation by using the above-mentioned modules. For details, please refer to the description of the above-mentioned related method embodiments, which will not be repeated here.

[0140] Figure 4 This is a schematic diagram according to the fourth embodiment of the present disclosure; the agent-based creative solution generation device 400 of this embodiment, in the above... Figure 3 Based on the technical solutions of the illustrated embodiments, the technical solutions of this disclosure will be described in further detail. For example... Figure 4 As shown, the agent-based creative solution generation device 400 of this embodiment includes the above-described... Figure 3 The modules with the same name and function shown are: input acquisition module 401, set acquisition module 402, generation module 403, and filtering module 404.

[0141] In this embodiment, the verification information set includes a reference scheme set, a verification dataset set for verifying creative schemes, and a verification index set.

[0142] Further optionally, in one embodiment of this disclosure, the set acquisition module 402 is used for:

[0143] Obtain the verification information set from the input information; or

[0144] Using the intelligent agent, the reference scheme set is generated based on the input information; using the intelligent agent, the verification dataset set and the verification index set are obtained based on the reference scheme set.

[0145] Further optionally, in one embodiment of this disclosure, the set acquisition module 402 is used for:

[0146] Using the aforementioned intelligent agent, based on the specified paper and the creative goal, multiple relevant reference solutions are retrieved from a specified database;

[0147] Based on the specified paper and the multiple related reference schemes, the reference scheme set is constructed.

[0148] Further optionally, in one embodiment of this disclosure, the set acquisition module 402 is used for:

[0149] Obtain an initial reference scheme from the input information;

[0150] Using the aforementioned intelligent agent, based on the initial reference scheme, the specified paper, and the creative goal, the agent retrieves the multiple relevant reference schemes from the specified database.

[0151] The set acquisition module 402 is also used for:

[0152] Based on the specified paper, the multiple related reference schemes, and the initial reference scheme, the reference scheme set is constructed.

[0153] Further optionally, in one embodiment of this disclosure, the generation module 403 includes:

[0154] The gap generation unit 4031 is used by the intelligent agent to generate gap information based on the reference information, each reference scheme in the reference scheme set and the creative goal, wherein the gap information is used to indicate the problems existing in achieving the creative goal;

[0155] The solution generation unit 4032 is used to generate the plurality of first candidate creative solutions by employing the intelligent agent based on the creative goal, the reference information, the gap information and each of the reference solutions in the reference solution set.

[0156] Further optionally, in one embodiment of this disclosure, the scheme generation unit 4032 is used for:

[0157] Using the intelligent agent, multiple initial creative solutions are generated based on the creative goal, the reference information, the gap information, and each of the reference solutions in the reference solution set;

[0158] Using the aforementioned intelligent agent, multiple first extended creative solutions are generated based on the multiple initial creative solutions;

[0159] Based on the multiple initial creative schemes and the multiple first extended creative schemes, the multiple first candidate creative schemes are generated.

[0160] Further optionally, in one embodiment of this disclosure, the scheme generation unit 4032 is used for:

[0161] The plurality of initial creative solutions and the plurality of first extended creative solutions are respectively used as alternative creative solutions to construct a set of alternative creative solutions;

[0162] Based on the set of verification information, each of the candidate creative solutions is comprehensively scored to obtain a first score for each of the candidate creative solutions;

[0163] Based on the first score of each of the candidate creative solutions, multiple candidate creative solutions with the first score greater than a first preset threshold are selected from the set of candidate creative solutions and used as the multiple first candidate creative solutions.

[0164] Further optionally, in one embodiment of this disclosure, the scheme generation unit 4032 is used for:

[0165] Using each reference scheme in the reference scheme set of the verification information set as a reference benchmark, and based on partial data in each verification dataset in the verification dataset set and the verification index set, each candidate creative scheme is comprehensively scored to obtain a first score value for each candidate creative scheme.

[0166] Further optionally, in one embodiment of this disclosure, the filtering module 404 is configured to:

[0167] Based on the set of verification information, each of the first candidate creative solutions is comprehensively scored to obtain a second score for each of the first candidate creative solutions.

[0168] Based on the second score of each of the first candidate creative schemes and the second preset threshold, the first target creative scheme is obtained from the plurality of first candidate creative schemes.

[0169] Further optionally, in one embodiment of this disclosure, the scheme generation unit 4032 is used for:

[0170] Using each reference scheme in the reference scheme set as a reference benchmark, and based on the full data in each verification dataset in the verification dataset set and the verification index set, a comprehensive score is given to each first candidate creative scheme to obtain a second score value for each first candidate creative scheme.

[0171] Further optionally, in one embodiment of this disclosure, the scheme generation unit 4032 is used for:

[0172] Using the full data in each of the verification datasets, the main verification indicators in the verification indicator set corresponding to each first candidate creative scheme and each reference scheme are verified to obtain the value of the main verification indicator when each of the verification datasets verifies each first candidate creative scheme and each reference scheme, as well as the verification runtime when each first candidate creative scheme and each reference scheme is verified.

[0173] Based on the values ​​of the main verification indicators obtained when verifying each first candidate creative scheme and each reference scheme using each verification dataset, the ranking score of each first candidate creative scheme is obtained.

[0174] Based on the verification runtime corresponding to each of the first candidate creative solutions and each reference solution, obtain the penalty score for each of the first candidate creative solutions;

[0175] For each of the first candidate creative solutions, the ranking score and the penalty score of the first candidate creative solution are weighted and subtracted to obtain the second score of the first candidate creative solution.

[0176] Further optionally, in one embodiment of this disclosure, the scheme generation unit 4032 is used for:

[0177] For each of the first candidate creative solutions, the ranking of the first candidate creative solution when using each of the verification datasets is determined based on the value of the main verification index obtained when verifying the first candidate creative solution and each reference solution using each of the verification datasets.

[0178] For each of the first candidate creative solutions, the average of the reciprocals of the rankings of the first candidate creative solutions during verification on each of the verification datasets is taken as the ranking score of the corresponding first candidate creative solution.

[0179] Further optionally, in one embodiment of this disclosure, the scheme generation unit 4032 is used for:

[0180] Obtain the maximum and minimum verification runtime among the verification runtimes of each of the reference schemes;

[0181] Based on the verification runtime of each of the first candidate creative solutions, as well as the maximum and minimum verification runtimes, the penalty score for each of the first candidate creative solutions is determined using the following formula:

[0182] Penalty(h) = max(0, (Time(h) - Time_min_base) / (Time_max_base -Time_min_base));

[0183] h is the first candidate creative solution, Penalty(h) is the penalty score of the first candidate creative solution h, Time_min_base is the minimum verification runtime, Time_max_base is the maximum verification runtime, and Time(h) is the verification runtime of the first candidate creative solution h.

[0184] Further optional, such as Figure 4 As shown, in one embodiment of this disclosure, the agent-based creative solution generation device 400 further includes:

[0185] The detection module 405 is used to periodically detect whether the reference scheme set needs to be updated using the intelligent agent.

[0186] The update module 406 is used to obtain the updated reference scheme when the reference scheme set needs to be updated, and to add the updated reference scheme to the reference scheme set.

[0187] Further, optionally, in one embodiment of this disclosure, the generation module 403 is also configured to:

[0188] Using the intelligent agent, based on the updated reference scheme and the first target creative scheme, a plurality of second extended creative schemes are generated;

[0189] Based on the first target creative solution and the multiple second extended creative solutions, multiple second candidate creative solutions are generated;

[0190] Based on the full data in each of the verification datasets in the verification dataset set and the verification index set, a comprehensive score is given to each of the second candidate creative solutions to obtain a score value for each of the second candidate creative solutions;

[0191] Based on the scores of each of the second candidate creative solutions, the second target creative solution is obtained.

[0192] The agent-based creative solution generation device 400 of this embodiment achieves the same implementation principle and technical effect as the agent-based creative solution generation by using the above-mentioned modules. For details, please refer to the description of the above-mentioned related method embodiments, which will not be repeated here.

[0193] The acquisition, storage, and application of any type of information, such as user personal information, involved in the technical solutions disclosed herein comply with relevant laws and regulations and do not violate public order and good morals.

[0194] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0195] Figure 5A schematic block diagram of an example electronic device 500 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0196] like Figure 5 As shown, device 500 includes a computing unit 501, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 502 or a computer program loaded from storage unit 508 into random access memory (RAM) 503. RAM 503 may also store various programs and data required for the operation of device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via bus 504. Input / output (I / O) interface 505 is also connected to bus 504.

[0197] Multiple components in device 500 are connected to I / O interface 505, including: input unit 506, such as keyboard, mouse, etc.; output unit 507, such as various types of monitors, speakers, etc.; storage unit 508, such as disk, optical disk, etc.; and communication unit 509, such as network card, modem, wireless transceiver, etc. Communication unit 509 allows device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0198] The computing unit 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as the methods of this disclosure. For example, in some embodiments, the methods of this disclosure may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the methods of this disclosure described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the methods of this disclosure by any other suitable means (e.g., by means of firmware).

[0199] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0200] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0201] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0202] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0203] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0204] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0205] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0206] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for generating creative solutions based on intelligent agents, comprising: Based on user input, obtain reference information and creative goals for generating creative solutions; The reference information includes at least one of a specified paper and a specified topic; Obtain a set of verification information for the creative solution; Using an intelligent agent, multiple first candidate creative solutions are generated based on the reference information, the creative goal, and the set of verification information; Based on the set of verification information, a first target creative solution is selected from the plurality of first candidate creative solutions.

2. The method according to claim 1, wherein, The verification information set includes a reference scheme set, a verification dataset set for verifying creative schemes, and a verification indicator set.

3. The method according to claim 2, wherein, Obtain a set of verification information for the creative solution, including: Obtain the verification information set from the input information; or Using the intelligent agent, the reference scheme set is generated based on the input information; using the intelligent agent, the verification dataset set and the verification index set are obtained based on the reference scheme set.

4. The method according to claim 3, wherein, When the reference information includes a specified paper, the intelligent agent generates the reference scheme set based on the input information, including: Using the aforementioned intelligent agent, based on the specified paper and the creative goal, multiple relevant reference solutions are retrieved from a specified database; Based on the specified paper and the multiple related reference schemes, the reference scheme set is constructed.

5. The method according to claim 4, wherein, Using the aforementioned intelligent agent, based on the specified paper and the creative goal, multiple relevant reference solutions are retrieved from a specified database, including: Obtain an initial reference scheme from the input information; Using the aforementioned intelligent agent, based on the initial reference scheme, the specified paper, and the creative goal, the agent retrieves the multiple relevant reference schemes from the specified database. Specifically, based on the specified paper and the multiple related reference schemes, the reference scheme set is constructed, including: Based on the specified paper, the multiple related reference schemes, and the initial reference scheme, the reference scheme set is constructed.

6. The method according to claim 2, wherein, Using an intelligent agent, based on the reference information, the creative goal, and the set of verification information, multiple first candidate creative solutions are generated, including: Using the aforementioned intelligent agent, gap information is generated based on the reference information, each reference scheme in the reference scheme set, and the creative goal. The gap information is used to indicate the problems existing in achieving the creative goal. Using the intelligent agent, the plurality of first candidate creative solutions are generated based on the creative goal, the reference information, the gap information, and each of the reference solutions in the reference solution set.

7. The method according to claim 6, wherein, Using the aforementioned intelligent agent, based on the creative objective, the reference information, the gap information, and each of the reference schemes in the reference scheme set, the plurality of first candidate creative schemes are generated, including: Using the intelligent agent, multiple initial creative solutions are generated based on the creative goal, the reference information, the gap information, and each of the reference solutions in the reference solution set; Using the aforementioned intelligent agent, multiple first extended creative solutions are generated based on the multiple initial creative solutions; Based on the multiple initial creative schemes and the multiple first extended creative schemes, the multiple first candidate creative schemes are generated.

8. The method according to claim 7, wherein, Based on the plurality of initial creative schemes and the plurality of first extended creative schemes, the plurality of first candidate creative schemes are generated, including: The plurality of initial creative solutions and the plurality of first extended creative solutions are respectively used as alternative creative solutions to construct a set of alternative creative solutions; Based on the set of verification information, each of the candidate creative solutions is comprehensively scored to obtain a first score for each of the candidate creative solutions; Based on the first score of each of the candidate creative solutions, multiple candidate creative solutions with the first score greater than a first preset threshold are selected from the set of candidate creative solutions and used as the multiple first candidate creative solutions.

9. The method according to claim 8, wherein, Based on the verification information set, each of the candidate creative solutions is comprehensively scored to obtain a first score for each candidate creative solution, including: Using each reference scheme in the reference scheme set of the verification information set as a reference benchmark, and based on partial data in each verification dataset in the verification dataset set and the verification index set, each candidate creative scheme is comprehensively scored to obtain a first score value for each candidate creative scheme.

10. The method according to claim 2, wherein, Based on the verification information set, a first target creative solution is selected from the plurality of first candidate creative solutions, including: Based on the set of verification information, each of the first candidate creative solutions is comprehensively scored to obtain a second score for each of the first candidate creative solutions. Based on the second score of each of the first candidate creative schemes and the second preset threshold, the first target creative scheme is obtained from the plurality of first candidate creative schemes.

11. The method according to claim 10, wherein, Based on the verification information set, each of the first candidate creative solutions is comprehensively scored to obtain a second score for each of the first candidate creative solutions, including: Using each reference scheme in the reference scheme set as a reference benchmark, and based on the full data in each verification dataset in the verification dataset set and the verification index set, a comprehensive score is given to each first candidate creative scheme to obtain a second score value for each first candidate creative scheme.

12. The method according to claim 11, wherein, Using each reference scheme in the reference scheme set of the verification dataset as a reference benchmark, and based on the full data in each verification dataset and the verification index set, a comprehensive score is given to each first candidate creative scheme to obtain a second score for each first candidate creative scheme, including: Using the full data in each of the verification datasets, the main verification indicators in the verification indicator set corresponding to each first candidate creative scheme and each reference scheme are verified to obtain the value of the main verification indicator when each of the verification datasets verifies each first candidate creative scheme and each reference scheme, as well as the verification runtime when each first candidate creative scheme and each reference scheme is verified. Based on the values ​​of the main verification indicators obtained when verifying each first candidate creative scheme and each reference scheme using each verification dataset, the ranking score of each first candidate creative scheme is obtained. Based on the verification runtime corresponding to each of the first candidate creative solutions and each reference solution, obtain the penalty score for each of the first candidate creative solutions; For each of the first candidate creative solutions, the ranking score and the penalty score of the first candidate creative solution are weighted and subtracted to obtain the second score of the first candidate creative solution.

13. The method according to claim 12, wherein, Based on the values ​​of the main verification metrics obtained when verifying each of the first candidate creative solutions and each reference solution using the respective verification datasets, the ranking score of each of the first candidate creative solutions is obtained, including: For each of the first candidate creative solutions, the ranking of the first candidate creative solution when using each of the verification datasets is determined based on the value of the main verification index obtained when verifying the first candidate creative solution and each reference solution using each of the verification datasets. For each of the first candidate creative solutions, the average of the reciprocals of the rankings of the first candidate creative solutions during verification on each of the verification datasets is taken as the ranking score of the corresponding first candidate creative solution.

14. The method according to claim 12, wherein, Based on the verification runtime corresponding to each of the first candidate creative solutions and each reference solution, a penalty score is obtained for each of the first candidate creative solutions, including: Obtain the maximum and minimum verification runtime among the verification runtimes of each of the reference schemes; Based on the verification runtime of each of the first candidate creative solutions, as well as the maximum and minimum verification runtimes, the penalty score for each of the first candidate creative solutions is determined using the following formula: Penalty(h) = max(0, (Time(h) - Time_min_base) / (Time_max_base - Time_min_base)); h is the first candidate creative solution, Penalty(h) is the penalty score of the first candidate creative solution h, Time_min_base is the minimum verification runtime, Time_max_base is the maximum verification runtime, and Time(h) is the verification runtime of the first candidate creative solution h.

15. The method according to any one of claims 2-14, wherein, Based on the verification information set, after selecting the first target creative solution from the plurality of first candidate creative solutions, the method further includes: The intelligent agent is used to periodically detect whether the reference scheme set needs to be updated; When the reference scheme set needs to be updated, the updated reference scheme is obtained; and the updated reference scheme is added to the reference scheme set.

16. The method according to claim 15, wherein, The method further includes: Using the intelligent agent, based on the updated reference scheme and the first target creative scheme, a plurality of second extended creative schemes are generated; Based on the first target creative solution and the multiple second extended creative solutions, multiple second candidate creative solutions are generated; Based on the full data in each of the verification datasets in the verification dataset set and the verification index set, a comprehensive score is given to each of the second candidate creative solutions to obtain a score value for each of the second candidate creative solutions; Based on the scores of each of the second candidate creative solutions, the second target creative solution is obtained.

17. An agent-based creative solution generation device, comprising: The input acquisition module is used to obtain reference information and creative goals for generating creative solutions based on user input information; The reference information includes at least one of a specified paper and a specified topic; The collection acquisition module is used to acquire a collection of verification information for validating creative solutions; The generation module is used to generate multiple first candidate creative solutions using an intelligent agent based on the reference information, the creative target, and the set of verification information. The filtering module is used to filter the first target creative scheme from the plurality of first candidate creative schemes based on the set of verification information.

18. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1-16.

19. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-16.

20. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1-16.

Images