Large model training method and device, function recommendation method and device, equipment and medium

By acquiring sample dialogue information and its corresponding positive and negative sample labels, and using the erroneous recommendation results generated by the initial large model itself to train the large model, the problem of insufficient accuracy of the large model in functional recommendation tasks is solved, and efficient preference alignment and accuracy improvement are achieved.

CN121882271APending Publication Date: 2026-04-17BAIDU COM TIMES TECH (BEIJING) CO LTD
0 Cites 1 Cited by

Patent Information

Application Number
CN202610018132.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-07
Publication Date
2026-04-17

AI Technical Summary

Technical Problem

Existing large models lack accuracy in function recommendation tasks and struggle to effectively correct erroneous tendencies, resulting in poor human-computer interaction performance.

Method used

By acquiring sample dialogue information and its corresponding positive and negative sample labels, the initial large model generates incorrect recommendation results as negative sample labels using its own sampling operation, constructs specific positive and negative sample pairs, and trains the initial large model to improve its accuracy in functional recommendation tasks.

Benefits of technology

It significantly improves the accuracy of large models in functional recommendation tasks, achieves efficient preference alignment, corrects the model's erroneous tendencies, and improves the smoothness of human-computer interaction.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a large model training method and device, a function recommendation method and device, equipment and a medium, and relates to the technical field of artificial intelligence, in particular to the technical fields of natural language processing, deep learning, large language models, information recommendation, intelligent agents, intelligent agents and the like. According to the implementation scheme, first sample data are obtained, the first sample data comprise sample dialogue information and corresponding positive sample labels and negative sample labels, and the positive sample labels indicate target function recommendation results determined according to the sample dialogue information; the negative sample label is a function recommendation result which is inconsistent with the positive sample label and is obtained by inputting sample dialogue information into an initial large model to execute sampling operation; and training the initial large model based on the first sample data to obtain a target large model.
Need to check novelty before this filing date? Find Prior Art