Introduces LLM-friendly logical database design via three operators (+A, +P, +R) that yield up to 4.2% gains in execution accuracy on BIRD-Union and Spider-Union benchmarks.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.DB 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Metadata Reasoner uses agentic LLM reasoning on metadata to select sufficient and minimal data sources, achieving 83.16% F1 on KramaBench and 85.5% F1 on noisy synthetic benchmarks while avoiding low-quality tables 99% of the time.
citing papers explorer
-
The Case for Text-to-SQL Friendly Logical Database Design
Introduces LLM-friendly logical database design via three operators (+A, +P, +R) that yield up to 4.2% gains in execution accuracy on BIRD-Union and Spider-Union benchmarks.
-
An Agentic Approach to Metadata Reasoning
Metadata Reasoner uses agentic LLM reasoning on metadata to select sufficient and minimal data sources, achieving 83.16% F1 on KramaBench and 85.5% F1 on noisy synthetic benchmarks while avoiding low-quality tables 99% of the time.