Model training method and device and text similarity determination method and device
A text similarity and model training technology, which is applied in character and pattern recognition, instruments, electrical digital data processing, etc., can solve the problem of model accuracy reduction
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0037] Through research on the current text similarity model, the applicant found that the training difficulty of the text similarity model mainly focuses on two aspects, one is the model training and model parameter adjustment for multiple text similarity annotations, the other is similar The robustness problem of degree metric definition.
[0038] Among them, regarding the model training and model parameter adjustment for multiple text similarity data annotations, the current text similarity annotation includes two text annotation methods, one is triple annotation form (text A, text P, text N), which satisfies The similarity between text A and text P is greater than that between text A and text N. Generally, the text similarity model is trained by constructing a triplet loss function. The other is 0-1 labeling form (text A, text B, label 0 / 1), 0 means that text A is not similar to text B, 1 means that text A is similar to text B, and generally trains text similarity based on...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More - R&D
- Intellectual Property
- Life Sciences
- Materials
- Tech Scout
- Unparalleled Data Quality
- Higher Quality Content
- 60% Fewer Hallucinations
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com



