Multi-Model Question Answering With Consistency Checking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing large models in intelligent question-answering scenarios often generate incorrect answers due to their limited performance and varying network structures, leading to inconsistent and inaccurate responses.
Innovation Solution
Integrate multiple large models to generate answering contents, utilize a second large model for consistency determination, and output consistent contents as the final answer, leveraging their reasoning capabilities to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single large model is used for question-answering, then the system is simple and fast, but the accuracy and reliability of answers deteriorate due to limited model performance
Solution Approach 1:
The patent combines multiple large models (first large models and second large model) into an integrated question-answering system. Multiple first large models generate different answering contents, which are then evaluated by a second large model for consistency. This merging of multiple models improves answer accuracy and reliability while maintaining manageable system complexity through structured architecture.
Solution Approach 2:
The second large model acts as an intermediary that evaluates and determines consistency between answering contents generated by multiple first large models. This intermediary component ensures quality control and selects consistent answers, thereby improving reliability without requiring direct comparison of all model outputs.
2Reliability
If multiple large models are integrated to improve answer accuracy, then the reliability improves, but the processing time and computational cost increase
Solution Approach 1:
The system pre-configures a large model set including multiple first large models and a second large model before processing questions. This preliminary setup allows for efficient query processing where pre-loaded models can immediately generate and evaluate answers without initialization delays, reducing processing time while maintaining high reliability.
Solution Approach 2:
The second large model provides feedback by determining consistency between answering contents generated by multiple first large models. This feedback mechanism enables the system to identify and select consistent answers, improving reliability while the feedback loop is optimized to minimize additional processing time through efficient consistency evaluation.
Data Source
AI summary
A method and an apparatus for question-answering, a related device, and a computer program product are provided. An answering content corresponding to question information is generated via first large models in a configured large model set. A second large model is pre-configured, a consistency of the answering contents generated by the respective first large models is detected using reasoning capability of the second large model, and a determination result of whether each pair of the answering contents in the answering content set is consistent is obtained. If the determination result indicates that at least one pair of answering contents is consistent, the consistent answering contents are outputted as a final answer. An error may exist in a single large model, but if at least one pair of answering contents is consistent, an accuracy rate of the consistent answering contents can be improved.