Skill Store System for Generative AI Task Orchestration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computing devices face diminished user experience and wasted resources when necessary skills for tasks are unavailable, leading to increased frustration and inefficiency in automated systems.
Innovation Solution
A skill store system that uses a generative large language model to orchestrate skills, allowing for the transmission of unsolved skill requests to a remote skill store for retrieval and execution, enabling the completion of tasks by sourcing skills from a global and centralized repository.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If skills are stored locally in the computing device, then skill access speed is improved, but skill variety and availability are limited
Solution Approach 1:
The skill storage system is segmented into local storage (for frequently accessed skills) and remote storage (for comprehensive skill repository). The computing device maintains a local skill store for fast access while connecting to a remote skill store for extensive skill variety, resolving the contradiction between speed and variety through spatial segmentation of storage resources.
Solution Approach 2:
A skill resolution component acts as an intermediary between the generative model and the skill stores. This intermediary manages the coordination between local and remote skill stores, determining which skills to fetch locally and which to retrieve remotely, thereby optimizing both access speed and skill variety utilization.
2Adaptability or versatility
If a remote skill store is accessed for all skills, then skill availability is improved, but system complexity and response time increase
Solution Approach 1:
The skill retrieval process is segmented into two stages: first querying the local skill store for immediately available skills, then only fetching additional skills from the remote skill store when necessary. This staged approach reduces system complexity by avoiding unnecessary remote communications while maintaining comprehensive skill availability.
Solution Approach 2:
Skills are pre-loaded into the local skill store before they are needed. The system maintains a cached copy of frequently used skills locally, so when a task requires such skills, they are already available and don't need to be fetched remotely, reducing response time and system complexity.
3Adaptability or versatility
If skills are fetched on-demand from remote store, then skill variety is improved, but time delay and computational resources are increased
Solution Approach 1:
The system performs preliminary actions by pre-loading and caching skills in the local store before they are actually needed for task execution. This proactive approach ensures that when skills are required, they are immediately available locally, eliminating retrieval delays and reducing overall time loss.
Solution Approach 2:
The skill retrieval strategy is dynamic rather than static. The system continuously adapts its behavior based on usage patterns, fetching and caching skills proactively when predicted to be needed, while avoiding unnecessary remote fetches. This dynamic approach optimizes both skill variety access and time efficiency.
Data Source
AI summary
Systems and methods for interfacing with a skill store are provided herein. In some examples, a task is processed, using a generate large model (GLM) to orchestrate skills for performing the task. The orchestrated skills include a plurality of skills related to the task. At least one skill in the orchestrated skills is determined to not be available to the GLM, and an indication corresponding to the at least one skill is transmitted to a remote skill store. The indication may be associated with descriptions of the at least one skill, based on which similarities may be determined for retrieving skills from the remote skill store. A remote skill is received from the remote skill score that corresponds to the transmitted indication, and the task is performed using the generative LLM. The generative LLM uses the remote skill to perform the task.