Search

CN-122020305-A - Task template recommendation method, device, medium, electronic equipment and program product

CN122020305ACN 122020305 ACN122020305 ACN 122020305ACN-122020305-A

Abstract

A task template recommending method, a device, a medium, electronic equipment and a program product relate to the technical field of computers, and are characterized in that an interactive interface is displayed, and a first task template is displayed in the interactive interface, wherein the first task template is obtained based on a trained template recommending model, the template recommending model is obtained based on object attribute information of an object and template information of a candidate second task template, the second task template is used for indicating an agent to execute tasks based on task description information corresponding to the second task template, the object attribute information and the template information of the task template can be fused to conduct task template recommending, and the relation between the object and a user can be effectively learned by the template recommending model, so that the template recommending can accurately meet the requirements of users. By combining object attribute information and template information, a proper task template can be recommended in an agent scene, so that task execution quality and overall interaction experience are improved.

Inventors

  • SHANG YU
  • NIU HAIBO

Assignees

  • 北京字跳网络技术有限公司

Dates

Publication Date
20260512
Application Date
20260129

Claims (11)

  1. 1. A task template recommendation method, comprising: displaying an interactive interface, wherein the interactive interface is used for the interaction between an object and an intelligent agent; And displaying a first task template in the interactive interface, wherein the first task template is obtained based on a trained template recommendation model, the template recommendation model is obtained based on object attribute information of an object and template information of a candidate second task template, and the second task template is used for indicating the intelligent agent to execute tasks based on task description information corresponding to the second task template.
  2. 2. The method of claim 1, wherein the trained template recommendation model is obtained by: And training the initial template recommendation model based on the object attribute information of the object, the template information of the candidate second task template and the interaction information between the object and the second task template, and obtaining the trained template recommendation model.
  3. 3. The method of claim 2, wherein training the initial template recommendation model based on the object attribute information of the object, the template information of the candidate second task template, and the interaction information between the object and the second task template to obtain the trained template recommendation model, comprises: Based on the interaction information between the object and the second task template, obtaining the interaction deviation degree of the object for the second task template; inputting object attribute information of an object and template information of the second task template into an initial template recommendation model to obtain an output result of the initial template recommendation model after reasoning; And taking the interaction deviation degree as a supervision signal of the initial template recommendation model, and iteratively updating model parameters of the initial template recommendation model in combination with the output result to obtain the template recommendation model after training.
  4. 4. A method according to claim 3, wherein the interaction information comprises an interaction between the object and the second task template, a number of interactions corresponding to the interaction, and a trigger time of the interaction; the obtaining the interaction bias degree of the object for the second task template based on the interaction information between the object and the second task template comprises the following steps: for each type of interaction action, obtaining action deflection degree corresponding to the interaction action based on interaction times corresponding to the interaction action and corresponding time attenuation coefficients, wherein the time attenuation coefficients are obtained based on triggering time of the interaction action; And obtaining the interaction deviation degree of the object aiming at the second task template based on the action deviation degree corresponding to each type of interaction action.
  5. 5. A method according to claim 3, wherein said iteratively updating model parameters of said initial template recommendation model in combination with said output result based on said supervisory signal of said initial template recommendation model with said degree of interaction bias to obtain said trained template recommendation model comprises: Determining a weighted binary cross entropy loss based on the interaction bias degree and the output result; Determining a sorting hinge loss based on the interaction bias degree and the output result; based on the weighted binary cross entropy loss and the sorting hinge loss, obtaining the interaction deviation degree and the total loss corresponding to the output result; And iteratively updating model parameters of the initial template recommendation model according to the total loss to obtain the trained template recommendation model.
  6. 6. The method of any of claims 1-5, wherein the trained template recommendation model is obtained by: Training a first template recommendation model based on object attribute information of an object, template information of a candidate second task template and interaction information between the object and the second task template to obtain a trained first template recommendation model; Storing model parameters corresponding to the trained first template recommendation model in a database; And obtaining model parameters corresponding to the trained first template recommendation model from the database, and loading the model parameters to a second template recommendation model deployed on line to obtain the trained template recommendation model.
  7. 7. The method of any one of claims 1-5, wherein the first task template is obtained by: searching a first task template corresponding to the object identification in a cache based on the object identification of the object; under the condition that the first task template is not found, responding to the fact that the object type of the object is a preset type, inputting the object identification into the trained template recommendation model, and obtaining the first task template output by the trained template recommendation model; and determining the first task template from the candidate second task templates through preset rules in response to the object type to which the object belongs being not a preset type.
  8. 8. A task template recommendation device, comprising: The first display module is configured to display an interactive interface, wherein the interactive interface is used for the interaction between the object and the intelligent agent; The second display module is configured to display a first task template in the interactive interface, wherein the first task template is obtained based on a trained template recommendation model, the template recommendation model is obtained based on object attribute information of an object and template information of a candidate second task template, and the second task template is used for indicating the agent to execute tasks based on task description information corresponding to the second task template.
  9. 9. A computer readable medium having stored thereon a computer program, wherein the computer program, when being executed by a processing device, implements the steps of the method of any of claims 1-7.
  10. 10. An electronic device, comprising: a storage device having a computer program stored thereon; Processing means for executing said computer program in said storage means to carry out the steps of the method of any one of claims 1-7.
  11. 11. A computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the steps of the method of any of claims 1-7.

Description

Task template recommendation method, device, medium, electronic equipment and program product Technical Field The technical scheme relates to the technical field of computers, in particular to a task template recommending method, a device, a medium, electronic equipment and a program product. Background With the continuous development of artificial intelligence, intelligent Agent (Agent) technology has also received more and more attention, and has become an important research topic in the field of artificial intelligence. The task template of the intelligent agent is used as a reusable task structural description and can be used for standardizing task intention, tool instructions and execution steps of the intelligent agent, so that the execution efficiency and consistency of the intelligent agent are remarkably improved. Therefore, how to effectively utilize the task template meeting the user requirements to process the task and ensure the task processing effect becomes a problem to be solved urgently. Disclosure of Invention This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. In a first aspect, a task template recommendation method is provided, including: displaying an interactive interface, wherein the interactive interface is used for the interaction between an object and an intelligent agent; And displaying a first task template in the interactive interface, wherein the first task template is obtained based on a trained template recommendation model, the template recommendation model is obtained based on object attribute information of an object and template information of a candidate second task template, and the second task template is used for indicating the intelligent agent to execute tasks based on task description information corresponding to the second task template. In a second aspect, a task template recommendation device is provided, including: The first display module is configured to display an interactive interface, wherein the interactive interface is used for the interaction between the object and the intelligent agent; The second display module is configured to display a first task template in the interactive interface, wherein the first task template is obtained based on a trained template recommendation model, the template recommendation model is obtained based on object attribute information of an object and template information of a candidate second task template, and the second task template is used for indicating the agent to execute tasks based on task description information corresponding to the second task template. In a third aspect, a computer readable medium is provided, on which a computer program is stored, wherein the computer program, when being executed by a processing device, carries out the steps of the method according to the first aspect. In a fourth aspect, there is provided an electronic device comprising: a storage device having a computer program stored thereon; Processing means for executing said computer program in said storage means to carry out the steps of the method according to the first aspect. In a fifth aspect, a computer program product is provided, comprising a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to the first aspect. According to the technical scheme, the interactive interface is displayed, the first task template is displayed in the interactive interface, the first task template is obtained based on the trained template recommendation model, the template recommendation model is obtained based on the object attribute information of the object and the template information of the candidate second task template, the second task template is used for indicating the intelligent agent to execute tasks based on the task description information corresponding to the second task template, the object attribute information and the template information of the task template can be fused to conduct task template recommendation, the template recommendation model can also effectively learn the relation between the object and the user, and the template recommendation can accurately meet the user requirements. By combining object attribute information and template information, a proper task template can be recommended in an agent scene, so that task execution quality and overall interaction experience are improved. Additional features and advantages of the technical solution will be set forth in the detailed description which follows. Drawings The above and other features, advantages and aspects of the present invention will become more apparent by reference to the following detailed description when taken i