Dynamic Query Planning with LLM-Generated Facets and Filters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional information retrieval systems face challenges in efficiently processing user queries due to the need for explicit facet and filter selections, ambiguity in user input interpretation, and scalability issues with generative artificial intelligence models, leading to irrelevant results and high computational latency.
Innovation Solution
A dynamic query planning system utilizing a large language model to automatically select functions for query execution based on user input, dynamically configuring prompts to generate a query execution plan that includes facets and filters without explicit user selection, and managing GAI model interactions to reduce latency and improve relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional information retrieval systems require explicit facet and filter selections, then query execution control is improved, but user input burden and operation complexity increase
Solution Approach 1:
The system performs self-service by automatically generating facets and filters based on query execution plans without requiring explicit user input. The planner component autonomously determines which facets and filters to apply, eliminating the need for users to manually select them while maintaining precise query execution control.
Solution Approach 2:
The system performs preliminary action by pre-generating multiple query execution plans with different facet and filter combinations before presenting them to the user. This allows the system to prepare execution strategies in advance, reducing the user's input burden while ensuring reliable query control through pre-evaluated plans.
2Measurement precision
If generative artificial intelligence models are used to interpret user input, then query understanding is improved, but computational latency increases
Solution Approach 1:
The system segments the query processing into distinct components: user input interpretation, query execution plan generation, and query result retrieval. By separating these functions and using specialized components for each, the system reduces overall latency while maintaining precise query understanding through the dedicated planner component.
Solution Approach 2:
The system applies partial action by using generative AI models selectively only for the interpretation phase rather than for entire query processing. This limited application of AI reduces computational latency while still improving query understanding where it is most needed.
3Reliability
If multiple query execution plans are generated, then query relevance is improved, but system complexity increases
Solution Approach 1:
The system implements dynamics by generating multiple query execution plans with different facet and filter combinations, allowing the most relevant plan to be selected based on query characteristics. This dynamic plan generation improves query relevance while the modular architecture manages system complexity through organized plan storage and selection mechanisms.
Data Source
AI summary
Embodiments of the disclosed technologies include receiving a first query including at least one first query term and configuring at least one prompt to cause a large language model to translate the at least one first query term into a set of functions that can be executed to obtain at least one second query term and generate and output a plan that is executable to create a modified version of the first query based on the at least one second query term. The plan is obtained by applying the large language model to the at least one prompt as configured. The plan is executed to determine the at least one second query term and create the modified version of the first query. The modified version of the first query is executed to provide, via the user interface, a response to the first query.