Prompt Intermediary System for Generative AI Adaptability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional large language models struggle with providing prompts in a flexible manner due to their trained states, limiting the ability to adapt to user-specific needs.
Innovation Solution
An information processing device that includes a reception unit to receive user setting information, a search unit to find appropriate prompts based on this information, and a providing unit to deliver these prompts to generative AI via an API, allowing for more flexible and user-specific prompt generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a large language model generates additional sentences based on a user's question, then the prompt is enriched with effective sentences, but the system lacks flexibility to adapt to different user settings and requirements
Solution Approach 1:
The patent introduces an intermediary system that sits between the user's question and the large language model. This intermediary receives user setting information, searches for appropriate prompts based on these settings, and then provides the customized prompt to the generative AI. This mediator layer enables flexibility and adaptability without requiring the LLM itself to be reconfigured, thus resolving the contradiction between adaptability and system complexity.
Solution Approach 2:
The system is divided into distinct functional modules: a reception unit for receiving user settings, a search unit for finding appropriate prompts, and a providing unit for delivering customized prompts. This segmentation allows each component to perform its specific function independently, making the system more adaptable to different user requirements while maintaining manageable complexity through modular design.
2Adaptability or versatility
If the large language model uses a fixed trained state, then the model structure remains stable, but the ability to provide user-specific prompts is limited
Solution Approach 1:
The system performs preliminary actions by receiving user setting information before generating the actual prompt. The search unit pre-searches for appropriate prompts based on user settings, and the providing unit prepares customized prompts in advance. This preliminary customization allows the system to adapt to user-specific needs while the large language model itself remains stable with its fixed trained state, resolving the contradiction between adaptability and reliability.
Data Source
AI summary
An information processing device according to the present application includes a reception unit, a search unit, and a providing unit. The reception unit receives, from another information processing device having received user setting information that is information set by a user, the user setting information via an API defined in advance. The search unit searches for a prompt to be input to generative AI that is used in the other information processing device, based on the user setting information received by the reception unit. The providing unit provides, as a provided prompt, the prompt searched by the search unit to the other information processing device via the API.