Dynamic Example Selection for Data Processing Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data processing methods rely on fixed example data, which may not adapt well to different types of natural language texts, leading to suboptimal processing of target request data in various task scenarios.
Innovation Solution
A data processing method that acquires target request data, determines a first example set from a preset library based on similarity calculations, and sends prompt information including the target request data and the first example set to a target model for contextual learning, allowing the model to process the data more accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If fixed example data is used for model processing, then the model structure remains simple, but the processing accuracy deteriorates for different types of natural language texts
Solution Approach 1:
The patent applies dynamics by transforming the static fixed example data into dynamic adaptive example selection. The system dynamically selects example data from a library based on similarity calculations between the target text and stored examples, allowing the processing approach to adapt to different input types without changing the model structure.
Solution Approach 2:
The patent changes the parameter of example data selection from fixed to variable. By introducing similarity thresholds and dynamic selection criteria, the system adjusts which examples are used based on the specific input, improving accuracy while maintaining model simplicity.
2Measurement precision
If dynamic example selection based on similarity is implemented, then processing accuracy for different text types improves, but system complexity increases
Solution Approach 1:
The patent segments the example data into a library of discrete examples with associated metadata. This segmentation allows efficient retrieval and selection based on similarity metrics, managing complexity through organized data structures rather than monolithic processing.
Solution Approach 2:
The patent introduces similarity calculation as an intermediary mechanism between the input text and example data selection. This mediator layer enables accurate matching without requiring complex model modifications, bridging the gap between simple model structure and adaptive processing.
3Measurement precision
If example data is selected to match target request data, then relevance and accuracy improve, but data processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-processing and storing example data in an organized library with extracted features and metadata before actual processing. This advance preparation enables faster similarity matching during runtime, reducing the time penalty of dynamic selection.
Solution Approach 2:
The patent applies local quality by focusing similarity calculations on specific relevant features rather than entire data sets. By identifying and comparing key characteristics locally, the system achieves accurate matching without processing all data comprehensively, reducing time consumption.
Data Source
AI summary
The present application provides a data processing method and apparatus, and a storage medium. Specifically, after target request data to be processed is acquired, a first example set may be determined from a preset example library according to the target request data. The first example set includes a plurality of pieces of first example data, and the first example data may provide a reference when the target request data is processed. Next, prompt information may be sent to a target model. The prompt information includes the target request data and the first example set. The target model may learn the first example set and process the target request data after learning the first example set. After completing processing the target request data, the target model may return an output result obtained based on the prompt information.